Direct interaction between Nrf2 and p21(Cip1/WAF1) upregulates the Nrf2-mediated antioxidant response

Abstract

In response to oxidative stress, Nrf2 and p21(Cip1/WAF1) are both upregulated to protect cells from oxidative damage. Nrf2 is constantly ubiquitinated by a Keap1 dimer that interacts with a weak-binding (29)DLG motif and a strong-binding (79)ETGE motif in Nrf2, resulting in degradation of Nrf2. Modification of the redox-sensitive cysteine residues on Keap1 disrupts the Keap1-(29)DLG binding, leading to diminished Nrf2 ubiquitination and activation of the antioxidant response. However, the underlying mechanism by which p21 protects cells from oxidative damage remains unclear. Here we present molecular and genetic evidence suggesting that the antioxidant function of p21 is mediated through activation of Nrf2 by stabilizing the Nrf2 protein. The (154)KRR motif in p21 directly interacts with the (29)DLG and (79)ETGE motifs in Nrf2 and thus competes with Keap1 for Nrf2 binding, compromising ubiquitination of Nrf2. Furthermore, the physiological significance of our findings was demonstrated in vivo using p21-deficient mice.

Figure 1. Nrf2 was Required for the p21-Dependent Cellular Protection in Response to Oxidative Stress

(A) p21 exhibited antioxidant functions. HCT116-p21^+/+, HCT116-p21^-/-, and HCT116-p21^-/- cells transfected with an expression vector for p21, were treated with the indicated doses of H₂O₂ for 12 hrs. Intracellular ROS levels were measured by the dichlorofluorescein/flow cytometry method. The representative histograms of ROS analysis in HCT116-p21^+/+cells and HCT116-p21^-/- cells are shown in the upper two panels. (B) HCT116-p21^+/+ cells with knockdown of Nrf2 had higher intracellular levels of ROS. HCT116-p21^+/+ cells were transfected with either control siRNA or Nrf2-siRNA. At 72 hrs post-transfection, levels of Nrf2 and intracellular ROS were measured. (C) Ectopic expression of p21 enhanced cell survival in response to H₂O₂ in MEF-Nrf2^+/+, but not in MEF-Nrf2^-/- cells. MEF-Nrf2^+/+ or MEF-Nrf2^-/- cells were cotransfected with the indicated expression vectors for 24 hrs. GFP was used to mark the transfected cells. Following treatment with H₂O₂, over 150 GFP positive cells were counted for both surviving and apoptotic cells by fluorescence microscopy. The percentage of surviving cells was plotted. (D) Knockdown of p21 sensitized cells to H₂O₂-induced cell death in MEF-Nrf2^+/+, but not in MEF-Nrf2^-/- cells. Cells were transfected with p21-siRNA or control siRNA for 72 hrs before treatment with H₂O₂. Apoptotic cells were detected using the Annexin V-FITC/flow cytometry method. All error bars indicate standard deviations calculated from three independent experiments. *p<0.05 compared with its control.

Figure 2. p21 Upregulated the Protein Level of Nrf2 and its Downstream Genes

(A) Knockdown of p21 diminished expression of the Nrf2-target genes in MEF-Nrf2^+/+ cells. MEF-Nrf2^+/+ cells were transfected with either control siRNA or p21-siRNA. The endogenous levels of p21 and Nrf2 proteins were detected by immunoblot analysis at 48 hrs post-transfection, while the mRNA expression of Nrf2, Keap1, NQO1 and HO-1 was measured 72 hrs after transfection. The error bars indicate standard deviations calculated from triplicate samples. *p<0.05 compared with its control. (B) Knockdown of p21 had no effects in MEF-Nrf2^-/- cells. The same experiment was performed in MEF-Nrf2^-/- cells as described in 2A. (C) p21 upregulated the ARE-reporter gene activity. HCT116-p21^-/- cells were transfected with different amounts of p21 expression vectors, along with expression vectors for the NQO1-ARE firefly luciferase and TK-renilla luciferase. The protein levels of p21 and Nrf2 were detected with anti-p21 and anti-Nrf2 antibodies (lower panels). The firefly luciferase activities were normalized to renilla luciferase activities and the standard deviations were calculated from three independent experiments, each with duplicate samples. *p<0.05 compared with its control.

Figure 3. HCT116-p21^-/- Cells had Reduced Basal and Induced Levels of the Nrf2-Dependent Antioxidant Response

(A) p21 was upregulated in response to oxidative stress. HCT116-p21^+/+ cells were either untreated or treated with the indicated compounds for 16 hrs. Endogenous p21 and β-actin were detected with anti-p21 and anti-β-actin antibodies (left panel). The band intensity of p21 and β-actin was quantified and the p21 protein level was normalized to the β-actin protein and plotted (right panel). (B) The basal and induced Nrf2 protein levels were lower in HCT116-p21^-/- cells. HCT116-p21^+/+ or HCT116-p21^-/-cells were either untreated or treated with different doses of tBHQ or SF for 16 hrs. Endogenous Nrf2 was detected with an anti-Nrf2 antibody. Nrf2 appears as a double band because a lower percentage gel (7.5%) was used. (C) Knockdown of p21 diminished the transcriptional activity of Nrf2. HCT116-p21^+/+ and HCT116-p21^-/- cells were transfected with expression vectors for NQO1-ARE firefly luciferase and TK-renilla luciferase. The transfected cells were treated with the indicated concentration of tBHQ or SF for 16 hrs prior to the measurement of luciferase activities. The standard deviations were calculated from three independent experiments, each with duplicate samples. *p<0.05 compared with its control. (D) mRNA expression of the Nrf2-downstream genes were reduced in HCT116-p21^-/- cells. HCT116-p21^+/+ and HCT116-p21^-/- cells were either left untreated, or treated with 50 μM tBHQ or 10 μM SF for 16 hrs. Relative amounts of NQO1, Gpx2, and Mrp2 mRNAs were measured by qRT-PCR. The standard deviations were calculated from triplicate samples. *p<0.05 compared with its control. (E) HCT116-p21^-/- cells had lower NQO1 activity. NQO1 enzyme activity in HCT116-p21^+/+ and HCT116-p21^-/- cells was measured as the dicoumarol-inhibitable fraction of DCPIP reduction. NQO1 activity was normalized to the total protein level and the standard deviations were calculated from triplicate samples. *p<0.05 compared with its control.

Figure 4. p21 Directly Interacted with Nrf2

(A) p21 interacted with Nrf2. COS-1 cell lysates coexpressing p21-Myc and HA-Nrf2 were immunoprecipitated with anti-HA (left panel) or anti-Myc (right panel) antibodies. The total lysates (lower two panels) and the immunoprecipitates (upper two panels) were subjected to immunoblot analysis with both anti-HA and anti-Myc antibodies for detection of Nrf2 and p21. (B) The C-terminal domain of p21 was required for direct interaction with Nrf2. GST pull-down analysis were performed with GST-p21 fusion proteins containing different regions of p21 and the ³⁵S-labeled Nrf2 protein. (C) p21 directly interacted with Nrf2. The GST-Nrf2 fusion protein was used to pulldown ³⁵S-labeled p21. Luciferase was used as a negative control. (D) The N-terminal domain of Nrf2 interacted with p21. p21 and each of the Nrf2 C-terminal deletion mutants were coexpressed in COS-1 cells. The Nrf2-containing complexes were immunoprecipitated with HA-beads and blotted with anti-HA and anti-Myc antibodies for detection of Nrf2 and p21. Bands labeled with asterisks are Nrf2 and its deletion mutants. (E) Binding of endogenous p21 to Nrf2 was enhanced in response to oxidative stress. Cell lysates from HCT116-p21^+/+cells untreated or treated with tBHQ were subjected to immunoprecipitation analysis with an anti-p21 antibody and then blotted with an anti-Nrf2 antibody (upper panel). An aliquot of total lysate was subjected to immunoblot analysis with anti-Nrf2, anti-p21 and anti-tubulin antibodies (lower three panels).

Figure 5. p21 Interfered with the Keap1-Dependent Ubiquitination of Nrf2 by Competing with Keap1 for Nrf2 Binding Mainly through the DLG Motif

(A) p21 interacted with the DLG and ETGE motifs of Nrf2. Two Nrf2 mutants were generated: mDLG and mETGE in which the DLG or ETGE motif was replaced with alanine residues respectively. COS-1 cells were either cotransfected with expression vectors for the indicated Nrf2 protein and p21-Myc; or expression vectors for the indicated Nrf2 protein and Kelch-Myc. Transfected cells were immunoprecipitated with an anti-Nrf2 antibody and the immunoprecipitated proteins were subjected to immunoblot analysis with anti-HA and anti-Myc antibodies (upper panels). An aliquot of total lysate was analyzed with anti-HA, anti-Myc, and anti-tubulin antibodies (lower panels). (B) p21 reduced ubiquitination of Nrf2. HCT116-p21^+/+ and HCT116-p21^-/- cells were either untreated or treated with 100 μM tBHQ for 4 hrs in the presence of 10 μM of MG132, which blocks degradation of the ubiquitinated proteins. Cell lysates were subjected to an in vivo ubiquitination assay for detection of the ubiquitin-conjugated endogenous Nrf2 protein. Lysates were denatured and immunoprecipitated with an anti-Nrf2 or an anti-HA antibody (negative control), and blotted with an anti-ubiquitin antibody (upper panel). An aliquot of total lysate was analyzed for Nrf2 and GAPDH expression (lower panels). (C) p21 stabilized Nrf2 in basal and stressed conditions. HCT116-p21^+/+ or HCT116-p21^-/- cells were left untreated or pre-treated with 100 μm tBHQ for 4 h, followed by addition of 25 μM CHX and incubated for the time periods indicated. Endogenous Nrf2 was detected and the intensity of the Nrf2 bands was quantified and plotted on a semi-log graph. The amount of Nrf2 before addition of CHX was set as 1.

Figure 6. The C-terminal ¹⁵⁴KRR in p21 was Essential for the Binding and Upregulation of Nrf2

(A) mKRR in p21 interacted with Nrf2 and was required for upregulation of Nrf2. Three p21 mutants were made. mRKRR, mKRR, and mKRK are mutants with RKRR, KRR, or KRK replaced with methionine residues, respectively. COS-1 cells were cotransfected with expression vectors for Nrf2 and the indicated p21 protein. Transfected cells were subjected to immunoprecipitation with an anti-HA antibody. The immunoprecipitated proteins and an aliquot of total lysate were electrophoresed and analyzed with anti-HA and anti-Myc antibodies. (B) mKRR was unable to upregulate the transcriptional activity of Nrf2. COS-1 cells were transfected with the expression vector for the indicated p21 protein, along with expression vectors for the NQO1-ARE firefly luciferase and TK-renilla luciferase. The reporter gene analysis was performed as described in Fig 2C. All error bars indicate standard deviations calculated from three independent experiments. *p<0.05 compared with wild-type. (C) Localization of endogenous Nrf2 and p21. Nrf2 and p21 in MDA-MB-231 cells were determined by double-label indirect immunofluorescence with anti-Nrf2 and anti-p21 antibodies. Colocalization of Nrf2 and p21 is indicated by the presence of yellow in the merge images. (D) Localization of p21 and its mutants. NIH 3T3 cells were cotransfected with HA-Nrf2 and each of the Myc-tagged p21 proteins. Double-label indirect immunofluorescence was performed with anti-HA and anti-Myc antibodies for detection of Nrf2 and p21, respectively.

Figure 7. p21-Deficient Mice had Reduced Basal and Induced Levels of Nrf2 and Nrf2 Target Genes

(A) p21-deficient mice had reduced basal and induced levels of Nrf2 and Nrf2 target genes. Wild-type or p21-deficient mice (n=3) were treated with 350 mg/kg BHA for 12 hrs through intraperitoneal injection. Liver tissues were subjected to immunoblot analysis with anti-Nrf2, anti-NQO1, anti-HO-1, and anti-GAPDH antibodies. (B) A model by which p21 upregulates the Nrf2-dependent antioxidant response at both basal and induced conditions. Keap1, an E3 ubiquitin ligase containing the BTB and Kelch domains, constantly targets Nrf2 for ubiquitination and subsequent degradation under basal conditions. The binding of Keap1 to Nrf2 is through a hinge and latch mechanism, in which each Kelch domain from a Keap1 homodimer binds Nrf2 through two binding sites: a weak-binding DLG motif and a strong-binding ETGE motif. Binding of both sites are essential to present the seven ubiquitin-accepting lysine residues of Nrf2 in the correct orientation to accept ubiquitin, which targets Nrf2 for degradation. In response to oxidative stress, Keap1 is able to detect an imbalance in intracellular redox homeostasis through modification of its cysteine residues. The three important cysteine residues, C151, C273, and C288 are labeled. These modifications alter conformation of Keap1 and loosen the latch (DLG-Kelch), which puts p21 in a better position to compete with Keap1 for binding to the DLG motif. Once the latch is open, the Keap1-dependent ubiquitination of Nrf2 is compromised even though Keap1 still associates with Nrf2 through the hinge (ETGE). Thus, Nrf2 is stabilized and Nrf2-dependent cytoprotective genes are expressed under oxidative stress.