Biliary excretion of arsenic by human HepaRG cells is stimulated by selenide and mediated by the multidrug resistance protein 2 (MRP2/ABCC2).

Millions of people worldwide are exposed to unacceptable levels of arsenic, a proven human carcinogen, in drinking water. In animal models, arsenic and selenium are mutually protective through formation and biliary excretion of seleno-bis (S-glutathionyl) arsinium ion [(GS)2AsSe]-. Selenium-deficient humans living in arsenic-endemic regions are at increased risk of arsenic-induced diseases, and may benefit from selenium supplementation. The influence of selenium on human arsenic hepatobiliary transport has not been studied using optimal human models. HepaRG cells, a surrogate for primary human hepatocytes, were used to investigate selenium (selenite, selenide, selenomethionine, and methylselenocysteine) effects on arsenic hepatobiliary transport. Arsenite + selenite and arsenite + selenide at different molar ratios revealed mutual toxicity antagonism, with the latter being higher. Significant levels of arsenic biliary excretion were detected with a biliary excretion index (BEI) of 14 ± 8%, which was stimulated to 32 ± 7% by selenide. Consistent with the formation and biliary efflux of [(GS)2AsSe]-, arsenite increased the BEI of selenide from 0% to 24 ± 5%. Arsenic biliary excretion was lost in the presence of selenite, selenomethionine, and methylselenocysteine. Sinusoidal export of arsenic was stimulated ∼1.6-fold by methylselenocysteine, but unchanged by other selenium forms. Arsenic canalicular and sinusoidal transport (±selenide) was temperature- and GSH-dependent and inhibited by MK571. Knockdown experiments revealed that multidrug resistance protein 2 (MRP2/ABCC2) accounted for all detectable biliary efflux of arsenic (±selenide). Overall, the chemical form of selenium and human MRP2 strongly influenced arsenic hepatobiliary transport, information critical for human selenium supplementation in arsenic-endemic regions.

Studies of selenium and arsenic mutual protection in human HepG2 cells.

Hundreds of millions of people worldwide are exposed to unacceptable levels of carcinogenic inorganic arsenic. Animal models have shown that selenium and arsenic are mutually protective through the formation and elimination of the seleno-bis(S-glutathionyl) arsinium ion [(GS)2AsSe]-. Consistent with this, human selenium deficiency in arsenic-endemic regions is associated with arsenic-induced disease, leading to the initiation of human selenium supplementation trials. In contrast to the protective effect observed in vivo, in vitro studies have suggested that selenite increases arsenite cellular retention and toxicity. This difference might be explained by the rapid conversion of selenite to selenide in vivo. In the current study, selenite did not protect the human hepatoma (HepG2) cell line against the toxicity of arsenite at equimolar concentrations, however selenide increased the IC50 by 2.3-fold. Cytotoxicity assays of arsenite + selenite and arsenite + selenide at different molar ratios revealed higher overall mutual antagonism of arsenite + selenide toxicity than arsenite + selenite. Despite this protective effect, in comparison to 75Se-selenite, HepG2 cells in suspension were at least 3-fold more efficient at accumulating selenium from reduced 75Se-selenide, and its accumulation was further increased by arsenite. X-ray fluorescence imaging of HepG2 cells also showed that arsenic accumulation, in the presence of selenide, was higher than in the presence of selenite. These results are consistent with a greater intracellular availability of selenide relative to selenite for protection against arsenite, and the formation and retention of a less toxic product, possibly [(GS)2AsSe]-.

Selenium-dependent and -independent transport of arsenic by the human multidrug resistance protein 2 (MRP2/ABCC2): implications for the mutual detoxification of arsenic and selenium.

Simultaneous exposure of lab animals to toxic doses of the human carcinogen arsenic (As) and the essential trace element selenium (Se) results in a remarkable mutual detoxification. A likely basis for this is the in vivo formation and biliary excretion of seleno-bis(S-glutathionyl) arsinium ion [(GS)(2)AsSe](-); however, the transport protein responsible for the biliary efflux of [(GS)(2)AsSe](-) has not been identified. The multidrug resistance protein 2 (MRP2/ABCC2) is an adenosine triphosphate (ATP)-binding cassette transporter expressed at the canalicular membrane of hepatocytes. Rat Mrp2 is known to excrete the As glutathione (GSH/GS-) conjugates arsenic triglutathione [As(GS)(3)] and monomethyl arsenic diglutathione [CH(3)As(GS)(2)] into bile, and in vitro studies have established As(GS)(3) as a substrate for human MRP2. In the present study, membrane vesicles prepared from human embryonic kidney (HEK293T) cells transfected with human MRP2 were used to demonstrate that MRP2 transports [(GS)(2)AsSe](-). In addition, the characteristics of MRP2 transport of As(GS)(3) and [(GS)(2)AsSe](-) were investigated. As(GS)(3) and [(GS)(2)AsSe](-) are chemically labile and have the potential to dissociate. However, arsenite (As(III)) +/- selenite (Se(IV)) transport was not detected in the absence of GSH or in the presence of the non-reducing GSH analog, ophthalmic acid, suggesting that the conjugates are the transported forms. The apparent K(m) values for [(GS)(2)AsSe](-) and As(GS)(3) were 1.7 and 4.2 microM, respectively, signifying high relative affinities. Membrane vesicles prepared from human erythrocytes, which express the MRP2-related MRP1/ABCC1, MRP4/ABCC4 and MRP5/ABCC5, transported As(GS)(3) in an MRP1- and ATP-dependent manner but did not transport [(GS)(2)AsSe](-). These results have important implications for the Se-dependent and -independent disposition of As.