Oxidative stress impairs the heat stress response and delays unfolded protein recovery

Abstract

Background: Environmental changes, air pollution and ozone depletion are increasing oxidative stress, and global warming threatens health by heat stress. We now face a high risk of simultaneous exposure to heat and oxidative stress. However, there have been few studies investigating their combined adverse effects on cell viability.

Principal findings: Pretreatment of hydrogen peroxide (H(2)O(2)) specifically and highly sensitized cells to heat stress, and enhanced loss of mitochondrial membrane potential. H(2)O(2) exposure impaired the HSP40/HSP70 induction as heat shock response (HSR) and the unfolded protein recovery, and enhanced eIF2alpha phosphorylation and/or XBP1 splicing, land marks of ER stress. These H(2)O(2)-mediated effects mimicked enhanced heat sensitivity in HSF1 knockdown or knockout cells. Importantly, thermal preconditioning blocked H(2)O(2)-mediated inhibitory effects on refolding activity and rescued HSF1 +/+ MEFs, but neither blocked the effects nor rescued HSF1 -/- MEFs. These data strongly suggest that inhibition of HSR and refolding activity is crucial for H(2)O(2)-mediated enhanced heat sensitivity.

Conclusions: H(2)O(2) blocks HSR and refolding activity under heat stress, thereby leading to insufficient quality control and enhancing ER stress. These uncontrolled stress responses may enhance cell death. Our data thus highlight oxidative stress as a crucial factor affecting heat tolerance.

Figure 1. Cell death in H₂O₂ and/or heat-treated T98G cells.

A) Trypan blue exclusion assay. Cell death 24 h after heat exposure (44°c for 20 min) (closed bars) with the indicated pretreatments (0.25 mM H₂O₂, 25 µg/ml VP16 or 50 nM FK228). Alternatively, cells were exposed to heat and thereafter treated with 0.25 mM H₂O₂ as reverse treatment (gray bar) or unexposed to heat (open bars). **, P<0.01 compared with cells treated alone as indicated. B) Cell death was evaluated 20 h after heat exposure. Cells were pretreated with 5 mM L-NAC, 10mM SB203580 or 10 mM SP600125 for 2 h, thereafter treated with 0.25 mM H₂O₂, and unexposed (open) or exposed (closed bars) to 44°c for 20 min. **, P<0.01 compared with heat/H₂O₂-treated cells. In A and B, error bars indicate the mean ± S.D. of data from three separate experiments. C) Disruption of Δψm. Cells were cultured for 20 h after the indicated treatments as described in A), and then intracellular DePsipher fluorescence was detected. Numbers indicate % of cells showing loss of Δψm. D) ROS generation. Carboxy-H₂DCFDA fluorescent signals 2 h after treatment with 0.25 mM H₂O₂ (H₂O₂), heat exposure (44°c for 20 min; Heat) or both (Both). Both (Rev) indicates reverse treatment (heat exposure before H₂O₂ treatment). In representative histograms, numbers indicate mean fluorescence intensity.

Figure 2. Effects of H₂O₂ on HSR and expression of scavenger enzymes.

A) Total RNA was harvested from T98G cells at the indicated hours after the indicated treatments (44°c for 20 min and/or 0.25 mM H₂O₂) and transcripts of the indicated genes were evaluated by RT-PCR. B) Confocal microscopical detection of HSP70. Cells were pretreated with 0.25 mM H₂O₂ (H₂O₂) or untreated (control) and fixed at 2 or 4 h after heat exposure. HSP70 expression was immuunohistochemically detected using the anti-HSP70 antibody and cells were counter-stained with propidium iodide (PI). Bar indicates 10 µm. C) eIF2α and JNK phosphorylation. At 1 and 5 h after indicated treatments as described above, eIF2α and JNK activities were evaluated by western blots using anti-phosphorylated eIF2α and anti-phosphorylated JNK antibodies, respectively. HSP70 and catalase expression levels were also evaluated. Anti-β-actin, anti-eIF2α and anti-JNK protein antibodies show equal loading of protein samples.

Figure 3. Effect of H₂O₂ on heat-exposed A172 cells.

A) Cell death at 24 h (upper) and disruption of Δψm (lower panel) at 20 h after heat (44°c for 20 min) exposure (closed bars) with or without 0.25 mM H₂O₂. Numbers indicate % of cells showing loss of Δψm. B) HSP70 transcripts. Total RNA was harvested from cells treated with the indicated and transcripts of the indicated genes were evaluated by RT-PCR (upper). HSP70 transcripts were determined by real-time PCR and normalized to GAPDH levels (lower panel). C) Time-course of HSP70 transcripts. At the indicated hours after heat exposure with (closed) or without (open bars) 0.25 mM H₂O₂, HSP70 transcripts were determined by real-time PCR and normalized to GAPDH levels In A-C, error bars indicate the mean ± S.D. of data from three separate experiments and **, P<0.01 compared with H₂O₂-untreated cells. D) Confocal microscopical detection of HSP70. Cells were pretreated with (H₂O₂) or without (control) 0.25 mM H₂O₂ and fixed at the indicated hours after heat (44°c for 20 min) exposure. HSP70 expression was immunohistochemically detected using the anti-HSP70 antibody counter-stained with PI. Bar indicates 10 µm. E) Bmf transcripts evaluated by RT-PCR of total RNA harvested at 6 h after treatment with the indicated agents for 6 h In B and E, GAPDH mRNA levels ensure that the RNA was correctly quantified.

Figure 4. ChIP assay and knockdown of HSF1 transcripts.

A) eIF2α phosphorylation in A172 cells. At the indicated hours after heat exposure (44°c for 20 min) with or without 0.25 mM H₂O₂ pretreatment, eIF2α activity was evaluated by western blots, and anti-eIF2α protein antibody shows equal loading of protein samples. B) Effect of H₂O₂ pretreatment on HSE transcription. A172 cells were transiently co-transfected with the pHSE and the pRL-TK reporter genes, pretreated with 0.25 mM H₂O₂ (closed bars), incubated for the indicated hours after heat exposure (43°c for 20 min), and their transcription activity was evaluated by dual luciferase assays. Bars display the mean ± S.D. of data from three separate experiments and *, P<0.05, **, P<0.01 compared with H₂O₂-unexposed cells. C) Chromatin immunoprecipitation assays. Cells (C; control) were incubated 2 h after heat exposure (44°c for 20 min) with (B; both) or without (H; heat alone) 0.25 mM H₂O₂ pretreatment. They were then fixed with formaldehyde, and immunoprecipitated with antibodies against HSF1 (anti-HSF1) or nonimmunized rabbit serum (serum). Immunoprecipitates were amplified by HSF1 promoter primers. A fixed portion of the total input was also examined by PCR (INPUT). D) HSF1 knockdown. Total cell lysates were harvested 36 h after transfection with siRandom (R) or two different siHSF1 (H1 and H2). E) XBP1 splicing. Total RNA was prepared from cells transfected with siRandom (R) or siHSF1 (H; H1). After transfection, cells were exposed to heat (44°c for 20 min) and incubated for the indicated hours. The indicated transcripts were evaluated by RT-PCR. GAPDH mRNA levels ensure that the RNA was correctly quantified. F) eIF2α phosphorylation. After exposure to heat (44.5°c for 20 min), eIF2α phosphorylation and HSP70 expression were evaluated by western blots. Anti-eIF2α protein antibody (in A and F) and anti-β-actin antibody (in D and F) show equal loading of protein samples.

Figure 5. Recovery of protein folding activity.

A) Refolding activity. Cells were transiently co-transfected with the pGRE/RL-TK reporter genes, treated with 5 mM dexamethazone for 10-12 h, exposed to heat (44°c for 20 min) with (closed) or without (open bars) 0.25 mM H₂O₂ pretreatment, incubated for 2 to 4 h and luciferase activity was measured. Cells were also incubated with 5 mM L-NAC for 30 min prior to 0.25 mM H₂O₂ pretreatment. *, P<0.05, **, P<0.01 compared with H₂O₂-untreated cells. B) HSP70 transcripts. Total RNA was harvested from cells treated as indicated and HSP70 transcripts were evaluated by RT-PCR (left). HSP70 transcripts were determined by real-time PCR and normalized to GAPDH levels (right panel). **, P<0.01 compared with heat/H₂O₂-treated cells. Columns display the mean ± S.D. of data from three separate experiments.

Figure 6. Effect of H₂O₂ on heat-exposed HSF +/+ and -/- MEFs.

A) Cell death at 24 h after heat exposure (42.5°c for the indicated min) with (closed) or without (open bars) 0.5 mM H₂O₂ pretreatment. B) HSP70 transcripts evaluated by RT-PCR transcription at the indicated hours after the indicated treatments (upper) were determined by real-time PCR with normalization to GAPDH levels (lower panel). C) Preconditioning effect on disruption of Δψm. Both MEF cells were treated as indicated, heat (Heat; 42.5°c for 20 min), 0.5 mM H₂O₂ treatment (H₂O₂) and both treatments (Both). As thermal preconditioning, cells were preheated (40.5°c for 30 min 10 h) prior to both treatments (Preheat/Both). Cells were cultured for 20 h after heat exposure and incubated with DePsipher solution. Numbers indicate % of cells showing loss of Δψm. D) Effect of H₂O₂ pretreatment on refolding activity. Refolding activity was evaluated by recovery of luciferase activity 5 h after heat exposure (42.5°c for 20 min) with (closed) or without (open bars) 0.5 mM H₂O₂ pretreatment. Cells were also treated by thermal preconditioning (Preheat) as described in C. In A, B and D, columns display the mean ± S.D. of data from three separate experiments and *, P<0.05; **, P<0.01 compared with H₂O₂-untreated cells.

Figure 7. Effect of thermal preconditioning on H₂O₂-treated HSF +/+ and -/- MEFs.

A) Cell death at 24 h after indicated treatments with (closed) or without (open bars) thermal preconditioning. Both MEFs were treated as indicated, heat (Heat; 42.5°c for 20 min), 0.5 mM H₂O₂ treatment (H₂O₂) and both treatments (Both). **, P<0.01 compared with un-preconditioned cells. B) HSF1 overexpression. The indicated HSF1 expression vectors were transfected into HSF1-/- MEFs and cell lysates were harvested at 2 days after transfection. β-actin demonstrates equal loading of protein samples (upper). Nuclear localization of HSF1 in HSF1-rescued MEFs was evaluated by confocal microscopy. The nucleus was counterstained with PI and bar indicates 10 µm. (lower panel). C) HSP70 transcripts were evaluated by RT-PCR transcription in cells 1 h after exposure to heat alone (H; 42.5°c 20 min), heat after 0.5 mM H₂O₂ pretreatment (B) or untreated cells (C). GAPDH mRNA levels ensure that the RNA was correctly quantified (upper). After indicated treatments, HSP70 transcripts in transfectants expressing wild-type (open) or mutant HSF1 (closed bars) were determined by real-time PCR and normalized to GAPDH levels (lower panel). **, P<0.01 compared with H₂O₂-untreated cells. In A and C, error bars indicate the mean ± S.D. of data from three separate experiments and

Figure 8. HSF1 rescue into HSF -/- MEFs.

A) Cell death at 24 h after heat exposure (42.5°c) for 20 min with (closed) or without (open bars) 0.5 mM H₂O₂ pretreatment. **, P<0.01 compared with MOCK transfectants. B) DNA fragmentation. Small molecular DNAs were prepared from cells (C: control, H: 42.5°c for 10 min, B: heat and 0.5 mM H₂O₂) 24 h after heat exposure. Numbers indicate molecular weights (kilo bases). C) Effect of H₂O₂ pretreatment on refolding activity. Cells were transiently co-transfected with the wild-type pHSF1 expression vector (HSF1) or its vehicle (MOCK) with the pGRE/RL-TK reporter genes, treated with 5 mM dexamethazone for 10-12 h, treated by thermal preconditioning as described in Fig. 6D , exposed to heat (42.5°c for 20 min) with (closed) or without (open bars) 0.5 mM H₂O₂ pretreatment, incubated for 5 h and luciferase activity was measured. **, P<0.01 compared with MOCK transfectants. D) eIF2α phosphorylation and XBP1 splicing. After exposure to heat (43.5°c for 20 min), HSF +/+ and -/- MEFs were harvested at the indicated hours. The eIF2α phosphorylation and HSP70 expression were evaluated by western blots using the indicated antibodies. Anti-β-actin antibody shows equal loading of protein samples. XBP1 splicing and HSP70 mRNA expression were evaluated by RT-PCR. GAPDH mRNA levels ensure that the RNA was correctly quantified. In A and C, error bars indicate the mean ± S.D. of data from three separate experiments.