TOX
Animal studies have demonstrated that the liver is the major target organ of Se toxicity (Diskin et al., 1979). In the present study, we found both Se forms caused liver injury, but SeMSC was more potent as evidenced by serum levels of ALT, AST, and LDH (Figs. 2A–C; Table 4) and the changes of liver architecture (Figs. 4B–D). Moreover, an oral dose of 5 mg Se/ kg for 7 days, SeMSC caused 20% mice death and suppressed mice growth earlier than Nano-Se (Fig. 4A); an oral dose of 10 mg Se/kg for 7 days, SeMSC and Nano-Se caused 80% and 10% mice death, respectively (Fig. 3). In addition, Nano-Se was 6-fold higher in LD50 value than SeMSC (Table 2). The mechanism for the toxic effects of Se has been suggested to occur due to its prooxidant ability to catalyze the oxidation of thiols and simultaneous generation of reactive oxygen species
(ROS), which can damage cellular components by lipid peroxidation(Spallholz,1994).MDAistheproductoflipidperoxidation caused by ROS. Both Se forms caused MDA increase, but its accumulation and duration in liver tissues of SeMSC-treated mice
was higher and longer than that for Nano-Se (Fig. 2D) which implied the extent of oxidative stress was different. In parallel to this, T-AOC, a sum of the activities of the various antioxidative substances, is significantly decreased by SeMSC but not by Nano-Se (Fig. 2F and Table 4); GST, responsible for detoxifying harmful compounds and scavenging lipid peroxidation (Hayes et al., 2005), was induced much earlier and higher by Nano-Se as compared with SeMSC (Fig. 2E and Table 4).
Empiric deduction suggests that Nano-Se has an impaired bioavailability due to the decrease in toxicity, especially to its
background of elemental Se, which has long been thought to be biologically inert. Therefore, a comparison of bioavailability was conducted by orally administering SeMSC and Nano-Se to Se-deficient mice once daily for 7 consecutive days. Our results herein showed that the investigated Se forms were equally able to increase GPx and TrxR (Fig. 6).
At nutritional levels of Se intake, Se accumulation was comparably efficient between the two Se forms (Fig. 5); however, at a supranutritional level the Se accumulation associated with Nano-Se was significantly lower compared with SeMSC in addition to plasma (Fig. 5). The Se species accumulated from SeMSC at supranutritional level may include methylselenol, an assumed reactive Se species involved in the anticarcinogenic effects of Se (Itoh and Suzuki, 1997). Multiple studies have demonstrated the ability of methylselenol to affect the cell cycle and induce apoptosis in cancer cell lines (Kim et
al., 2001;
al., 2001). Therefore, the potential higher accumulation of methylselenol by SeMSC at supranutritional level may lead to argue for its advantage over Nano-Se for chemoprevention. Although the actual chemopreventive mechanisms of Se are still not fully understood, selenoenzymes, phase 2 enzymes, and the cytotoxic effect of Se are all likely to be involved. Therefore, the presumed advantage of SeMSC over Nano-Se is largely dependent on the assumption of a selective modulation of cancer cells by Se and a predominant position occupied by cytotoxic pathway over other mechanistic pathways, such as selenoenzymes and phase 2 enzymes.
Effect of SeMSC and Nano-Se on GST activity. Se-deficient mice were orally administered SeMSC and Nano-Se at the dose of 500, 1000, and 1500 lg Se/kg once daily for 7 consecutive days. Activity of ALT in serum (A). Activity of AST in serum (B). Activity of GST in liver (C).
ZHANG, WANG, AND XU
Regarding selectivity, several investigators have reported that Se compounds preferentially inhibit growth and induce apoptosis in cancer cells compared with normal cells (Ghose
et
al., 2001; Watrach et
al., 1984). However, surprisingly, in the case of liver, the major target of Se toxicity in animals (Diskin et al., 1979), it was found that the cytotoxic effect of Se was at least equally or even more potent in normal hepatocytes compared with hepatic carcinoma cells (Weiller et al., 2004). Consistent with this study, in the case of prostate, an extensively investigated tissue in Se chemoprevention, nontumorigenic prostate cells are highly sensitive to Se toxicity as compared with prostate cancer cells at physiologically relevant concentrations (5–10lM) With respect to the rank of cytotoxic effect among the multiple pathways for chemopreventive effect of Se, to the best of our knowledge, no such a direct comparison has been conducted up to now. On the contrary, in animal studies of Se at supranutritional levels, the modulation of phase 2 enzymes has frequently been reported (El-Sayed et al., 2006; El-Sayed and Franklin, 2006; Ip and Lisk, 1997) and the involvement of selenoenzymes at nutritional levels has again received considerable attention (On the other hand, if the cytotoxic effect of Se is indeed the central mechanism or sole mechanism for its chemopreventive capacity, the risk of Se toxicity seems to be unavoidable unless a superior selectivity to cancer cells is firmly established because ingested supranutritional Se is overwhelmingly taken by normal cells compared with malignant cells in a chemopreventive setting, whereas the converse selectivity, opposite to expectation, that was reported by Weiller et
one cannot attribute the predominant chemopreventive mechanism of Se to cytotoxicity alone. In addition, many selenocompounds, such as 1,4-phenylenebis(methylene)selenocyanate and 2-substituted selenazolidine-4(R)-carboxylic acids, whose chemical structures are less likely to be metabolized into methylselenol, still possess convincing chemopreventive effects. A prominent profile of these selenocompounds is their capacity to induce phase 2 enzymes in animals at a level lower than the toxic dose (El-Sayed and Franklin, 2006; Sohn et al., 1999). Induction of phase 2 enzymes is an effective and sufficient strategy for achieving protection against the toxic and neoplastic effects of many carcinogens (Talalay, 2000). Therefore, if taking the toxicity of Se into account, the induction of phase 2 response might represent a more attractive mechanism compared with the cytotoxic mechanism.
Recently, methylselenol precursors were identified to be highly effective selenocompounds for the induction of phase 2 enzymes in Hepa 1c1c7 cells (Xiao and Parkin, 2006). It should be noted that the Se level needed for the induction of phase 2 enzymes can be much lower than those causing cell growth inhibition or cytotoxicity (Xiao and Parkin, 2006). As stated by Xiao and Parkin (2006), methylselenol
precursors were highly effective inducers for phase 2 response in cells; these authors suggested that animal studies were needed to assess the gap between effective dose and toxic dose. Our present study demonstrates that SeMSC dose dependently increases hepatic GST activity at supranutritional levels (0.5–1.5 mg Se/kg, daily, for 7 days) without causing toxicity, and that there is no significant difference between SeMSC and Nano-Se at equi-Se doses (Fig. 7). However, further escalated doses of SeMSC (5–10 mg Se/kg, daily, for 7 days) produce more prominent toxicity as compared with Nano-Se.
Significantly lower accumulation of Se at supranutritional levels in mice fed Nano-Se compared with SeMSC is relevant to the size effect of Nano-Se and is a strategy for avoiding toxicity by application of nanotechnology. It has been reported that Nano-Se has a size-dependent effect in scavenging various free radicals; small-size Nano-Se has greater ability to transfer electrons to radicals (Huang et al., 2003). However, the size effect of Nano-Se within 200 nm in terms of induction of selenoenzymes could not be observed in both Se-deficient cultured cells and Se-deficient mice at nutritional level (Zhang et al., 2004). The explanation to this phenomenon has two aspects: intrinsically, an active absorption pathway as driving force to obtain Se may exist when Se-deficient cells are thirsty for Se; extrinsically, even modest Se supplementation can easily saturate selenoenzymes, thereby excluding a size effect under an urgent need. At the supranutritional level, in fact, approaching to the toxic level, whereupon selenoenzymes have already been fully saturated, cells may in turn change to passive absorption of Se. Our unpublished results which compared Se retention of 36 and 90 nm Nano-Se at supranutritional levels showed large size constituted a barrier for Se accumulation. Compared with small molecular weight selenocompounds, for example, SeMSC herein, 36 nm Nano-Se is fairly larger in size. With respect to Se accumulation, SeMSC and 36 nm Nano-Se did not show a difference at nutritional levels, but did show a significant difference at supranutritional levels, which is exactlythesameastherulesinsizecomparisonbetween36and 90 nm Nano-Se. Reduced accumulation of Se at supranutritional levels may effectively delay the onset and development of Se toxicity; fortunately, the reduction of Se accumulation does not lead to attenuate the induction of GST. In summary, based on chemoprevention-related responses in animals, as compared with SeMSC, Nano-Se possesses equal efficacy in increasing the activity of GPx, TrxR, and GST, but has much lower toxicity, suggesting that Nano-Se can serve as a potential chemopreventive agent with significantly reduced risk of Se toxicity. FUNDING
This work was supported by a grant from University of
Science and Technology of
ACKNOWLEDGMENTS
The authors are grateful to Prof. Ethan Will Taylor for his help in
the revision of the manuscript. We also express our thanks to Hua-Li Wang,
Dun-geng Peng, Hong-juan Lu, and Gui-Lan He for technical assistance in the
experiments.
REFERENCES
Albrecht, M. A., Evans, C. W., and
Raston, C. L. (2006). Green chemistry and the health implications of
nanoparticles. Green Chem.
8, 417–432. Behne, D., and Kyriakopoulos, A. (1993). Effects of dietary
selenium on the tissue
concentrations of type I iodothyronine 5#-deiodinase and other selenoproteins. Am. J. Clin.
Nutr. 57, 310S–312S. Benzie,