Induction of ubiquitin C (UBC) gene transcription is mediated by HSF1: role of proteotoxic and oxidative stress

Abstract

The polyubiquitin gene ubiquitin C (UBC) is considered a stress protective gene and is upregulated under various stressful conditions, which is probably a consequence of an increased demand for ubiquitin in order to remove toxic misfolded proteins. We previously identified heat shock elements (HSEs) within the UBC promoter, which are responsible for heat shock factor (HSF)1-driven induction of the UBC gene and are activated by proteotoxic stress. Here, we determined the molecular players driving the UBC gene transcriptional response to arsenite treatment, mainly addressing the role of the nuclear factor-erythroid 2-related factor 2 (Nrf2)-mediated antioxidant pathway. Exposure of HeLa cells to arsenite caused a time-dependent increase of UBC mRNA, while cell viability and proteasome activity were not affected. Nuclear accumulation of HSF1 and Nrf2 transcription factors was detected upon both arsenite and MG132 treatment, while HSF2 nuclear levels increased in MG132-treated cells. Notably, siRNA-mediated knockdown of Nrf2 did not reduce UBC transcription under either basal or stressful conditions, but significantly impaired the constitutive and inducible expression of well-known antioxidant response element-dependent genes. A chromatin immunoprecipitation assay consistently failed to detect Nrf2 binding to the UBC promoter sequence. By contrast, depletion of HSF1, but not HSF2, significantly compromised stress-induced UBC expression. Critically, HSF1-mediated UBC trans-activation upon arsenite exposure relies on transcription factor binding to previously mapped distal HSEs, as demonstrated to occur under proteasome inhibition. These data highlight HSF1 as the pivotal transcription factor that translates different stress signals into UBC gene transcriptional induction.

Keywords: HSF1; Nrf2; UBC gene; proteostasis; stress response; ubiquitin upregulation.

Figure 1

Arsenite and proteasome inhibitor MG132 induce UBC gene expression and promote nuclear accumulation of HSF1, HSF2 and Nrf2 transcription factors. (A) HeLa cells were treated with NaAsO₂ (80 μm) or MG132 (20 μm) over an 8‐h time course (as indicated). UBC mRNA levels were determined by RT‐qPCR, normalized to GAPDH levels and expressed as fold increase relative to control cells (NT/DMSO). (B) HeLa cells treated for 8 h with different concentrations of NaAsO₂ (as indicated) were analyzed as in (A). UBC mRNA level is shown as fold increase versus the untreated sample (NT). (C) Cell viability of HeLa cells exposed to NaAsO₂ or MG132 for the indicated times (n = 3 each) were evaluated by the MTS assay and shown as a percentage with respect to the untreated‐cell value (time 0 h), set equal to 100. (D) Cells treated with NaAsO₂ or MG132 for 4 and 8 h and untreated cells (NT) were subjected to the proteasome activity assay (n = 4). (E) Nuclear levels of HSF1, HSF2 and Nrf2 transcription factors upon cell treatment with stressors for the indicated times were analyzed by western immunoblotting of nuclear fractions, with anti‐HSF1, anti‐HSF2 and anti‐Nrf2 specific antibodies. TFIID and Sp3 were used as nuclear loading controls. Arrows indicate the detected protein. (F) HSF1, HSF2 and Nrf2 mRNA levels in cells treated for 8 h with NaAsO₂ or MG132 (n = 3 each) were measured by RT‐qPCR and expressed as a fold change relative to NT cells. All data are expressed as means ± SEM of the indicated number of samples. Asterisks denote statistical significance, calculated by one‐way ANOVA, versus control as specified or indicated by bars; *P < 0.05; **P < 0.01; ***, ###P < 0.001; n.s., not significant.

Figure 2

siRNA‐mediated knockdown of HSF1, HSF2 and Nrf2 transcription factors. (A) HeLa cells were transiently transfected with control (GFP), HSF1, HSF2 or Nrf2 siRNAs (indicated as siGFP, siHSF1, siHSF2, siNrf2). At 48 h after transfection the mRNA levels of knockdown genes were measured by quantitative real‐time PCR. The results are normalized to GAPDH mRNA and depicted as fold change compared with siGFP transfected cells. Data are means ± SEM (n = 11). Asterisks denote statistical significance, calculated by Student's t test, versus GFP control siRNA; ***P < 0.001. (B) Western immunoblot of total proteins from HeLa cells transfected with the indicated siRNA, at 48 h post‐transfection. Equal amounts of total cellular proteins were loaded and probed with antibodies specific for HSF1, HSF2 and Nrf2. Blot was reprobed with anti‐actin as a loading control. (C) Nrf2 expression (relative to actin) in cells transfected with control GFP siRNA or siRNA targeting Nrf2, after 8 h treatment with 20 μm MG132. Arrows indicate the detected protein; ns, not specific.

Figure 3

UBC gene induction by oxidative and proteotoxic stress is independent of Nrf2. (A) HeLa cells receiving control GFP siRNA (siGFP) or Nrf2 specific siRNA (siNrf2) were treated, 48 h after transfection, with 80 μm NaAsO₂ or 20 μm MG132 for 8 h or left untreated as control (NT). UBC,GCLC and HMOX1 mRNAs were measured by RT‐qPCR, normalized to GAPDH levels and expressed as fold increase relative to untreated siGFP‐transfected cells (n = 6). (B) ChIP‐PCR analysis of HeLa cells before (NT) or after 4 h exposure to NaAsO₂ or MG132 (same concentration as in (A)). Chromatin was immunoprecipitated with Nrf2 specific antibody or with non‐specific anti‐rabbit IgG (as internal IP control). RT‐qPCR was performed on the immunoprecipitated samples (IP Nrf2 and IP ns IgG) and on the chromatin before immunoprecipitation (Input), using primers that amplify three different UBC promoter regions (FR1 −916/−759; FR5 −278/−187; FR6 −196/−96). Amplification with primers specific for an ARE‐containing locus in the NQO1 promoter served as positive control. Binding is depicted as the percentage of input values and is the mean of three independent experiments. Data are means ± SEM of the indicated experiments. Statistical analysis was performed by Student's t test (A) or one‐way ANOVA (B). Asterisks denote statistical significance (as indicated by bars) *P < 0.05; **P < 0.01; ***P < 0.001; n.s., not significant. #P < 0.05 and ###P < 0.001 indicate statistical significance versus NQO1_ARE NT sample.

Figure 4

Downmodulation of Nrf2 does not impair UBC gene induction by oxidative and proteotoxic stress in the mouse cell line NIH3T3. (A) NIH3T3 cells were transiently transfected with Nrf2 targeting siRNA and control GFP siRNA. At 48 h after transfection Nrf2 mRNA levels were measured by RT‐qPCR, normalized to GAPDH mRNA and depicted as fold change versus GFP siRNA transfected cells (n = 5). (B) siNrf2 and siGFP transfected cells were subjected to 8 h treatment with 80 μm NaAsO₂ or 20 μm MG132 or left untreated (NT) and Nrf2 protein levels were determined by immunoblot analysis of nuclear extracts with anti‐Nrf2 specific antibody. Sp3 was stained as nuclear loading control. Arrows indicate the detected protein. (C) UBC mRNA as well as GCLC and HMOX1 mRNAs were measured by RT‐qPCR, in NIH3T3 cells treated as indicated. The mRNA levels of analyzed genes were normalized to GAPDH levels and expressed as fold change relative to untreated siGFP transfected cells (n = 5). Data shown in (A,C) are the means ± SEM from the indicated number of samples. Asterisks denote statistical significance, calculated by Student's t test, versus control as indicated by bars; *P < 0.05; **P < 0.01; ***P < 0.001; n.s., not significant.

Figure 5

UBC gene induction by oxidative and proteotoxic stress is dependent on HSF1. (A,B) HeLa cells were transfected with either HSF1 and HSF2 specific siRNAs or control GFP siRNA. Forty‐eight hours after transfection, cells were treated with 80 μm NaAsO₂ or 20 μm MG132 for 8 h or left not treated as control (NT). UBC and HSP70 mRNAs were measured by RT‐qPCR, normalized to GAPDH levels and expressed as fold increase relative to not treated siGFP‐transfected cells. Data are means ± SEM of six experiments; asterisks denote statistical significance, calculated by Student's t test (as indicated by bars): *P < 0.05; **P < 0.01; n.s., not significant. (C) Expression analysis of promoter luciferase‐reporter constructs was performed in HeLa cells transiently transfected with the wild‐type construct P1 and its mutant derivatives (P1 mut FR1‐2 and P1 mut FR6) and treated, 48 h post‐transfection, with 80 μm NaAsO₂ for 8 h or with no stressor added. Luciferase mRNA was detected by RT‐qPCR and expression data, normalized to GAPDH, were compared with the value obtained for the untreated wild‐type construct (P1, NT) (n = 6). Bars indicate significant differences between P1 and P1 mutant‐driven luciferase expression, upon arsenite exposure, calculated by one‐way ANOVA. ***P < 0.001; n.s., not significant.