A Critical Role for Cysteine 57 in the Biological Functions of Selenium Binding Protein-1

Abstract

The concentration of selenium-binding protein1 (SBP1) is often lower in tumors than in the corresponding tissue and lower levels have been associated with poor clinical outcomes. SBP1 binds tightly selenium although what role selenium plays in its biological functions remains unknown. Previous studies indicated that cysteine 57 is the most likely candidate amino acid for selenium binding. In order to investigate the role of cysteine 57 in SBP1, this amino acid was altered to a glycine and the mutated protein was expressed in human cancer cells. The SBP1 half-life, as well as the cellular response to selenite cytotoxicity, was altered by this change. The ectopic expression of SBP1(GLY) also caused mitochondrial damage in HCT116 cells. Taken together, these results indicated that cysteine 57 is a critical determinant of SBP1 function and may play a significant role in mitochondrial function.

Keywords: glutathione peroxidase 1; mitochondrial damage; selenite cytotoxicity; selenium-binding protein1.

Figure 1

SBP1^GLY was impaired in its ability to protect HCT116 cells from selenite-induced toxicity. (A) Codon 57 of SBP1 was changed from a cysteine to a glycine and expression constructs representing the native and mutated SBP1 were introduced into HCT116 cells that do not produce detectable levels of SBP1. SBP1 signals were quantified and normalized to β-actin. Relative intensities were indicated; (B) HCT116 cells transfected with vector, pIRES2-SBP1, or pIRES2-SBP1^GLY were treated with 10 µM sodium selenite for 48 h and then photoed under microscope. Scale bar = 200 µm; and (C) cell number was quantified using ImageJSoftware. Results were analyzed with Student’s t-test and shown as mean ± SD. ** p < 0.01.

Figure 2

GPx1 extends endogenous SBP1 protein half-life. After treated with 100 µg/mL cycloheximide for 24, 48, and 72 h, different MCF7 derivative cells were harvested and protein levels of SBP1 were determined by immunoblot analysis. The signals were quantified by densitometry using ImageJ Software and normalized to β-actin. The determined half-life of SBP1 was 45, 64, and 62 h in MCF7, MCF7^A5P, and MCF7^A7L cells, respectively. The differences between MCF7 and MCF7^A5P, MCF7^A7L cells are statistically significant (p < 0.05).

Figure 3

SBP1^GLY has a shorter protein half-life than SBP1. HCT116 cells transfected with vector, pIRES2-SBP1, or pIRES2-SBP1^GLY were treated with 100 µg/mL cycloheximide for 24, 48 and 72 h and then protein levels of SBP1 were determined by immunoblot analysis. The signals were quantified by densitometry using ImageJ Software and normalized to β-actin. The determined half-life of SBP1 and SBP1^GLY were 63 and 45 h in HCT116 cells which has a statistically significant difference (p < 0.05).

Figure 4

Differential effects of wild-type SBP1 and SBP1^GLY on signaling pathways relevant to carcinogenesis. (A–C) After 48 h of transfection of HCT116 cells with vector, pIRES2-SBP1, or pIRES2-SBP1C57G, protein levels including phospho-p53, p53, phospho-PDK1, PDK1, GSK3β, phospho-GSK3β, total AKT, phospho-AKT (serine 473), PTEN, and cytochrome C were examined by immunoblot analysis. SBP1 signals were quantified and normalized to β-actin. Relative intensities were indicated.

Figure 5

SBP1^GLY induced mitochondrial damage in HCT116 cells. Mitochondrial damage in HCT116 cells expressing either SBP1 and SBP1^GLY, or the empty vector, were examined by using nonyl-acradine orange which assesses mitochondrial damage by staining cardiolipinin the inner mitochondrial membrane. Relative fluorescence units quantified using excitation and emission filters at 580 nm and 630 nm were reverse-related with cell mitochondrial damage. Results were analyzed with Student’s t-test and shown as mean ± SD. * p < 0.05.