Hydrogen sulfide inhibits the translational expression of hypoxia-inducible factor-1α

Abstract

Background and purpose: The accumulation of hypoxia-inducible factor-1α (HIF-1α) is under the influence of hydrogen sulfide (H(2) S), which regulates hypoxia responses. The regulation of HIF-1α accumulation by H(2) S has been shown, but the mechanisms for this effect are largely elusive and controversial. This study aimed at addressing the controversial mechanisms for and the functional importance of the interaction of H(2) S and HIF-1α protein.

Experimental approach: HIF-1α protein levels and HIF-1α transcriptional activity were detected by Western blotting and luciferase assay. The mechanisms for H(2) S-regulated HIF-1α protein levels were determined using short interfering RNA transfection, co-immunoprecipitation and 7-methyl-GTP sepharose 4B pull-down assay. Angiogenic activity was evaluated using tube formation assay in EA.hy926 cells.

Key results: The accumulation of HIF-1α protein under hypoxia (1% O(2) ) or hypoxia-mimetic conditions was reversed by sodium hydrosulfide (NaHS). This effect of NaHS was not altered after blocking the ubiquitin-proteasomal pathway for HIF-1α degradation; however, blockade of protein translation with cycloheximide abolished the effect of NaHS on the half-life of HIF-1α protein. Knockdown of eukaryotic translation initiation factor 2α (eIF2α) suppressed the effect of NaHS on HIF-1α protein accumulation under hypoxia. NaHS inhibited the expression of VEGF under hypoxia. It also decreased in vitro capillary tube formation and cell proliferation of EA.hy926 cells under hypoxia, but stimulated the tube formation under normoxia.

Conclusions and implications: H(2) S suppresses HIF-1α translation by enhancing eIF2α phosphorylation under hypoxia. The interaction of H(2) S and HIF-1α inhibits the angiogenic activity of vascular endothelial cells under hypoxia through the down-regulation of VEGF.

Figure 1

The effects of H₂S on HIF-1α protein levels in HEK293T cells challenged by hypoxia or hypoxia-mimetic agents. (A) HEK293T cells were treated under hypoxia for 4 h in the presence or absence of increasing concentrations of NaHS. The lower panel shown is the densitometric quantification of HIF-1α protein normalized to β-actin, and presented as % of hypoxia-alone group. *P < 0.05 versus hypoxia-alone group, n= 5. (B) HEK293T cells were co-transfected with plasmid HRE-luciferase and pRL-TK vector as described in Methods. After 24 h of transfection, cells were exposed to hypoxia for 4 h with or without 100 µM NaHS. Luciferase activity was measured and normalized to control. Each column represents the mean ± SEM (n= 3). *P < 0.05 versus group without NaHS. (C) HEK293T cells were transfected with plasmid HA-HIF1α-pcDNA3 for 24 h, followed by the treatment of 100 µM NaHS for 4 h under normoxic conditions. (D) HEK293T cells were co-transfected with plasmid HRE-luciferase and pRL-TK vector, and either with plasmid HA-HIF1α-pcDNA3 or empty vector. After 24 h of transfection, cells were challenged with 100 µM NaHS under normoxia. Luciferase activity was measured and normalized to control. Each column represents the mean ± SEM (n= 3). *P < 0.05 versus group without NaHS. (E) HEK293T cells were treated with either 200 µM DFX or 1 mM DMOG for 2 h, followed by the addition of 100 µM NaHS for 4 h. (F) HEK293T cells were treated with 200 µM DFX in either normoxic or hypoxic conditions for 4 h, with or without 100 µM NaHS. In (A, C, E and F), total protein extracts (40 µg) were subjected to immunoblot assays with anti-HIF-1α or anti-β-actin antibodies.

Figure 2

H₂S reduced HIF-1α protein levels in HEK293T cells in a time-dependent manner. (A) HEK293T cells were treated with 200 µM DFX for 4 h followed by 100 µM NaHS treatment for the indicated time. The levels of HIF-1α and β-actin were analysed by Western blotting. (B) HEK293T cells transfected with plasmid HA-HIF1α-pcDNA3 were treated with or without 100 µM NaHS for the indicated time under normoxic conditions. The levels of HIF-1α and β-actin were analysed by Western blotting. (C and D) HEK293T cells were treated with 200 µM DFX for 4 h followed by 100 µM NaHS treatment for the indicated time. The right panel shown is the densitometric quantification of HIF-1α protein normalized to β-actin, and presented as % of cells at 1 h (C) or 8 h (D) without NaHS. In all panels, mean ± SEM is shown, *P < 0.05 versus group without NaHS at each time point. n= 4.

Figure 3

The inhibitory effect of H₂S on HIF-1α protein level was independent of the ubiquitin-proteasomal degradation pathway. (A) Non-transfected HEK293T cells were treated with or without 100 µM NaHS in the presence of 20 µM MG132 for 4 h under either normoxia or hypoxia (left). Plasmid HA-HIF1α-pcDNA3 transfected HEK293T cells were exposed to the same treatments under normoxia (right). Total cell extracts (40 µg) were separated by SDS-PAGE and immunoblotted for HIF-1α protein; β-actin served as a loading control. (B) HEK293T cells co-transfected with plasmid HA-HIF1α-pcDNA3 and plasmid HA-ubiquitin were exposed to 20 µM MG132 with or without 100 µM NaHS for 4 h. Cell extracts were subjected to immunoprecipitation with antibody to ubiquitin (IP) followed by immunoblotting with anti-HIF-1α antibody. An aliquot of cell lysates that was reserved before IP was also analysed (input) by using anti-HIF-1α antibody. Ub, ubiquitin; Ub-HIF-1α, ubiquitinated HIF-1α. (C) HEK293T cells were treated with hypoxia for 4 h followed by re-oxygenation for the indicated time with or without 100 µM NaHS. Total cell extracts (40 µg) were prepared for immunoblotting. HEK293T cells were treated with either 1 mM DMOG (D) or 200 µM DFX (E) for 4 h, followed by the indicated treatments. The concentrations of CHX and NaHS are 25 and 100 µM respectively. Total cell extracts (40 µg) were prepared for immunoblotting with anti-HIF-1α antibody. The lower panels of (D) and (E) are the densitometric quantifications of HIF-1α normalized to β-actin, and presented as % of cells at 0 min. n= 3 for each group. Reoxy, re-oxygenation.

Figure 4

H₂S-induced HIF-1α down-regulation was independent of the mTOR/4E-BP1 pathway. (A) Western blot analysis of p53, p21 and cyclin D1 in HEK293T cells treated under hypoxia for 4 h with or without 100 µM NaHS; β-actin served as a loading control. (B) HEK293T cells were pretreated with or without 100 nM rapamycin for 0.5 h followed by exposure to hypoxia (upper) or 200 µM DFX (lower) for 4 h in the presence or absence of 100 µM NaHS respectively. Western blot analyses of HIF-1α and β-actin are shown. (C) HEK293T cells were exposed to hypoxia for 4 h with or without 100 µM NaHS. Cell lysates (500 µg) were subjected to 7-methyl GTP Sepharose, followed by immunoblotting with antibodies for eIF4G, 4E-BP1 and eIF4E respectively. Equal amounts of proteins that were reserved before pull down were also analysed (input).

Figure 5

Phosphorylation of eIF2α was associated with H₂S-induced HIF-1α translational repression in HEK293T cells. HEK293T cells were treated with 100 µM NaHS under normoxia (A) or hypoxia (B) for the indicated time. p-eIF2α were determined by Western blot analyses. The panel shown under (A) is the densitometric quantification of p-eIF2α normalized to total eIF2α, and presented as % of value at 0 min. *P < 0.05 versus 0 min, n= 5. Total eIF2α served as a loading control. HEK293T cells were transfected with a siRNA specific for eIF2α or a negative control siRNA. Transfected cells were exposed to hypoxia (C) or 200 µM DFX (D) for 4 h in the presence or absence of 100 µM NaHS. Levels of HIF-1α and eIF2α were examined by immunoblotting and β-actin served as a loading control. The middle and lower panels shown are the densitometric quantification of HIF-1α normalized to β-actin and eIF2α normalized to β-actin, respectively, and presented as % of negative control without NaHS from five (C) or seven (D) independent experiments. *P < 0.05 versus the same siRNA transfection group without NaHS. #P < 0.05 versus negative control group without NaHS. P < 0.05 versus negative control group without NaHS.

Figure 6

H₂S inhibited in vitro angiogenic activity in EA.hy926 cells under hypoxia. (A) EA.hy926 cells were incubated in hypoxic conditions for 4 h in the presence of NaHS at indicated concentrations. The right panel shown is the densitometric quantification of HIF-1α protein normalized to β-actin, and presented as % of hypoxia alone group. *P < 0.05 versus hypoxia-alone group, n= 4. (B) EA.hy926 cells were transfected with a siRNA specific for eIF2α or a negative control siRNA. Transfected cells were exposed to hypoxia for 4 h in the presence or absence of 100 µM NaHS. Levels of HIF-1α and eIF2α were examined by immunoblotting and β-actin served as a loading control. (C) EA.hy926 cells were exposed to hypoxia for 8 h with or without 100 µM NaHS. Total RNA was isolated. VEGF mRNA was analysed by real-time quantitative PCR. Each column represents the mean ± SEM (n= 3). *P < 0.05 versus hypoxic group without NaHS. (D) EA.hy926 cells were seeded onto 96-well plates coated with Matrigel, followed by incubation with 100 µM NaHS for 8 h under normoxia or hypoxia. 50 ng·mL⁻¹ VEGF was used as a positive control. Images of tube formation were taken under an inverted light microscope and tube branch points from each of four randomly chosen fields were quantified. Data represent the mean ± SEM, n= 3. *P < 0.05 versus control under normoxia. #P < 0.05 versus control under hypoxia. Cell viability (E) and cell proliferation (F) in EA.hy926 cells were detected under hypoxia in the presence of 100 µM NaHS for 24 h. *P < 0.05 versus hypoxia-alone group, n= 5.