NRF2 and Thioredoxin Reductase 1 as Modulators of Interactions between Zinc and Selenium

Abstract

Background: Selenium and zinc are essential trace elements known to regulate cellular processes including redox homeostasis. During inflammation, circulating selenium and zinc concentrations are reduced in parallel, but underlying mechanisms are unknown. Accordingly, we modulated the zinc and selenium supply of HepG2 cells to study their relationship.

Methods: HepG2 cells were supplied with selenite in combination with a short- or long-term zinc treatment to investigate intracellular concentrations of selenium and zinc together with biomarkers describing their status. In addition, the activation of the redox-sensitive transcription factor NRF2 was analyzed.

Results: Zinc not only increased the nuclear translocation of NRF2 after 2 to 6 h but also enhanced the intracellular selenium content after 72 h, when the cells were exposed to both trace elements. In parallel, the activity and expression of the selenoprotein thioredoxin reductase 1 (TXNRD1) increased, while the gene expression of other selenoproteins remained unaffected or was even downregulated. The zinc effects on the selenium concentration and TXNRD activity were reduced in cells with stable NRF2 knockdown in comparison to control cells.

Conclusions: This indicates a functional role of NRF2 in mediating the zinc/selenium crosstalk and provides an explanation for the observed unidirectional behavior of selenium and zinc.

Keywords: NRF2; selenium; selenium transporter; thioredoxin reductase; zinc.

Figure 1

Establishment of culture conditions with 2.5% FBS. HepG2 cells were treated with or without 50 nM sodium selenite in combination with or without 100 μM zinc sulfate for 72 h in media containing 10% or 2.5% FCS. The relative cell count (a) was determined by trypan blue exclusion test. Intracellular trace element concentrations of Se (b), and Zn (c) were determined by total reflection X-ray fluorescence spectrometry (TXRF). The measurement was performed for 1000 s with 1 mg/L yttrium as an internal standard. Protein expression of MT (d) was determined by Western blot, normalized to Ponceau staining. Either the −Se/−Zn 10% FBS group (a) or the +Se/+Zn group 10% FBS (b–d) were set as 1. Results are presented as mean + SD (n = 3–4). *** p < 0.001 vs. −Se, # p < 0.05, ## p < 0.01, ### p < 0.001 vs. −Zn, $$ p < 0.01, $$$ p < 0.001 vs. 10% FBS, calculated by two-factorial ANOVA with Bonferroni’s post-test.

Figure 2

Zinc increased the activity of TXNRD but not of GPX. HepG2 cells were treated with or without 50 nM sodium selenite in combination with or without 100 µM zinc sulfate for up to 72 h (a–e,g,h) or with indicated Zn concentrations (f) in media containing 2.5% FCS. Enzyme activities of GPX (a) and TXNRD (e,f) and the protein expression levels of GPX1 (b), GPX2 (c), GPX4 (d), TXNRD1 (g), and TXNRD2 (h) were determined photometrically (a,e,f) or by Western blot (b–d,g,h). Protein expression was normalized to Ponceau staining and presented relative to +Se/+Zn treatment. Results are presented as mean + SD (n = 3–4). * p < 0.05, *** p < 0.001 vs. −Se, # p < 0.05, ## p < 0.01, ### p < 0.001 vs. −Zn, calculated by two-factorial ANOVA (a–e,g,h) or with one-factorial ANOVA (f) with Bonferroni’s post-test.

Figure 3

Zinc enhanced the nuclear translocation of NRF2 and MTF1 and increased the expression of their target genes. HepG2 cells were treated with or without 50 nM sodium selenite in combination with or without 100 μM zinc sulfate up to 72 h (c,d) or treated with selenite for 72 h in combination with or without zinc for the time as indicated (a,b,e–j) in media containing 2.5% FCS. Nuclear protein levels of MTF1 (a), NRF2 (b) and SELENOH (g), enzyme activity of NQO1 (c), and protein expression of NQO1 (d) were determined photometrically (c) or by Western blot (a,b,d,g). Protein expression was normalized to Ponceau staining and presented relative to samples with selenium treatment in combination with 2 h zinc treatment (a,b) or to samples with selenium treatment in combination with 6 h zinc treatment (g) or samples with selenium treatment (d). The mRNA expression levels of NRF2 and MTF1 target genes (e–f,h–j) were analyzed by qPCR. Gene expression was normalized to the normalization factor of the reference genes HPRT, RPL13a, and GAPDH and presented relative to samples with selenium treatment and 6 h zinc treatment. Results are presented as mean + SD (n = 3–4). * p < 0.05, ** p < 0.01, *** p < 0.001 vs. −Se, # p < 0.05, ## p < 0.01, ### p < 0.001 vs. −Zn, $$$ p < 0.001 vs. 6 h Zn, calculated by two-factorial ANOVA with Bonferroni’s post-test.

Figure 4

NRF2 partially mediated the zinc effects on selenium homeostasis. HepG2 cells with a stable small hairpin RNA-mediated NRF2 knockdown (NRF2-KD) and scramble (Scr) control cells were treated with 50 nM sodium selenite for 72 h and 100 μM zinc sulfate for the indicated time (a–c) or for 72 h (d,e) in media containing 2.5% FCS. The nuclear protein levels of NRF2 (a,c) and MTF1 (b,c) were analyzed by Western blot. Protein expression was normalized to the Ponceau staining. Enzyme activities of TNXRD (d) and intracellular Se concentrations (e) were determined photometrically or by total reflection X-ray fluorescence spectrometry, respectively. The measurement was performed for 1000 s with 1 mg/L yttrium as an internal standard. The results are presented as mean + SD (n = 3). ### p < 0.001 vs. –Zn, *** p < 0.001 vs. 6 h Zn, $$$ p < 0.001 vs. Scr (a,b) or *** p < 0.001 vs. –Se, ### p < 0.001 vs. –Zn, $$ p < 0.01, $$$ p < 0.001 vs. Scr (d,e) calculated by two-factorial ANOVA with Bonferroni’s post-test.

Figure 5

Trace element transporters were increased by zinc. HepG2 cells were treated with or without 50 nM sodium selenite for 72 h in combination with or without 100 μM zinc sulfate for indicated time (up to 72 h) in media containing 2.5% FCS. The mRNA expression levels of ZIP8 (a), XCT (b), and APOER2 (d) were analyzed by qPCR. Gene expression was normalized to the normalization factor of the reference genes HPRT, RPL13a, and GAPDH and was presented relative to samples with selenium and 6 h zinc treatment. XCT (c) and APOER2 (e) were analyzed by Western blot. Protein expression was normalized to Ponceau staining and presented relative to samples with selenium and 48 h (c) or 72 h zinc treatment (e). Results are presented as mean + SD (n = 3–4). * p < 0.05 vs. −Se; # p < 0.05, ## p < 0.01, ### p < 0.001 vs. −Zn, $$$ p < 0.001 vs. 6 h Zn, calculated by two-factorial ANOVA (a,b,d) or with one-factorial ANOVA (c,e) with Bonferroni’s post-test.