Transcriptional Regulation of Selenoprotein F by Heat Shock Factor 1 during Selenium Supplementation and Stress Response

Abstract

Changes of Selenoprotein F (SELENOF) protein levels have been reported during selenium supplementation, stressful, and pathological conditions. However, the mechanisms of how these external factors regulate SELENOF gene expression are largely unknown. In this study, HEK293T cells were chosen as an in vitro model. The 5'-flanking regions of SELENOF were analyzed for promoter features. Dual-Glo Luciferase assays were used to detect promoter activities. Putative binding sites of Heat Shock Factor 1 (HSF1) were predicted in silico and the associations were further proved by chromatin immunoprecipitation (ChIP) assay. Selenate and tunicamycin (Tm) treatment were used to induce SELENOF up-regulation. The fold changes in SELENOF expression and other relative proteins were analyzed by Q-PCR and western blot. Our results showed that selenate and Tm treatment up-regulated SELENOF at mRNA and protein levels. SELENOF 5'-flanking regions from -818 to -248 were identified as core positive regulatory element regions. Four putative HSF1 binding sites were predicted in regions from -1430 to -248, and six out of seven primers detected positive results in ChIP assay. HSF1 over-expression and heat shock activation increased the promoter activities, and mRNA and protein levels of SELENOF. Over-expression and knockdown of HSF1 showed transcriptional regulation effects on SELENOF during selenate and Tm treatment. In conclusion, HSF1 was discovered as one of the transcription factors that were associated with SELENOF 5'-flanking regions and mediated the up-regulation of SELENOF during selenate and Tm treatment. Our work has provided experimental data for the molecular mechanism of SELENOF gene regulation, as well as uncovered the involvement of HSF1 in selenotranscriptomic for the first time.

Keywords: ER stress; Selenoprotein F; heat shock; heat shock factor 1; selenium supplementation; transcription regulation.

Figure 1

Selenoprotein F (SELENOF) mRNA up-regulation by selenate and tunicamycin (Tm). The transcription levels of SELENOF at several time intervals were quantified by Q-PCR in HEK293T cells (a) treated with selenate, (b) preconditioned with selenate then treated with Actinomycin D, (c) treated with Tm, and (d) preconditioned with Tm then treated with Actinomycin D. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 2

Identification of SELENOF regulatory elements. (a) Analysis of SELENOF (formerly known as SEP15) 5′-flanking regions in the Database of Transcriptional Start Sites (DBTSS). The SELENOF CDS was labeled with arrow and the CpG island was located up-stream of CDS. (b) Schematic locations of deletion derivatives designed for identification of SELENOF regulatory elements. Luciferase reporter activities of the deletion derivatives in HEK293T cells (c) and Neuro-2a cells (d). The empty vector pGL4.10 was used as the negative control. The data are representative of three independent experiments. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 3

Heat Shock Factor 1 (HSF1) increased SELENOF transcriptional regulation. (a) Luciferase reporter activities of SELENOF core positive regulatory element, (b) transcriptional levels of SELENOF, protein levels of (c) SELENOF and (d) GPX4 in control and HSF1 over-expressed HEK293T cells. The empty vector pcDNA3.1 was used as the negative control to be compared with pcDNA3.1-HSF1. In Luciferase reporter assays, the empty vector pGL4.10 was used as the negative control, while the pGL4-818/-248 was used as core promoter of SELENOF. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 4

HSF1 interacted with the 5′-flanking regions of SELENOF. (a) PCR amplification results of chromatin immunoprecipitation (ChIP) primers after protein-chromatin cross-linkage and immunoprecipitation. Locations of JASPAR predicted HSF1 binding sites and the designed ChIP primers were presented in the schematic map. (b) Manually searched results of both typical and less conserved variant of HSF1 Sequence-binding Elements (HSEs) in the sequences of SELENOF 5′- flanking region from −1430 to −248. The data are representative of three independent experiments.

Figure 5

Heat shock treatment increased SELENOF transcriptional regulation. (a) Luciferase reporter activities of SELENOF core positive regulatory element in transfected HEK293T cells with/without heat shock treatment. Transcriptional levels of SELENOF (b) and heat shock protein (HSP)70 (c) at several time intervals in HEK293T cells under heat shock treatment. (d) Protein levels of SELENOF in control and HSF1 over-expressed HEK293T cells with/without heat shock treatment. The data are representative of three independent experiments. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 6

The effect of selenate and Tm treatment on the endogenous HSF1 activation and SELENOF protein levels. (a) Protein levels of SELENOF, p-HSF1 (S326), and HSF1 at several time intervals in HEK293T cells treated with selenate. (b) Protein levels of Binding immunoglobulin protein (BIP), SELENOF, p-HSF1 (S326), and HSF1 at several time intervals in HEK293T cells treated with Tm. BIP bands were used as protein markers for adaptive endoplasmic reticulum (ER) stress. The data are representative of three independent experiments. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 7

HSF1 further increased SELENOF transcriptional regulation during selenate and Tm treatment. Luciferase reporter activities of SELENOF core positive regulatory element in control and HSF1 over-expressed HEK293T cells with/without (a) selenate and (b) Tm treatment. * p < 0.05, ** p < 0.01 and *** p < 0.001.

Figure 8

Involvement of HSF1 in the up-regulation of SELENOF during selenate and Tm treatment. Protein levels of SELENOF in (a) control and HSF1 over-expressed HEK293T cells with/without selenate treatment, (b) control and HSF1 over-expressed HEK293T cells with/without Tm treatment, (c) siRNA transfected HEK293T cells with/without selenate treatment, and (d) siRNA transfected HEK293T cells with/without Tm treatment. The data are representative of three independent experiments. The sham siRNAs were used as negative control to be compared with the siHSF1 siRNAs. * p < 0.05, ** p < 0.01 and *** p < 0.001.