Factors in fish modifying methylmercury toxicity and metabolism.

We report here some results of a long-term (19 month) study with cats fed methylmercury (MeHg) in nutritionally balanced diets based on fish. By using either freshwater pike (low in Se) or canned tuna (high in Se) as the major protein source, basal diets with low levels of MeHg were prepared having different Se content, all Se being of natural origin. The basal diets produced no signs of toxicity or pathological changes over the l9-month period. In cats fed basal diets spiked with medium or high levels of MeHg, evidence for delayed onset of toxic effects from the added MeHg was observed with the tuna diets compared to pike diets. In brain, muscle, and blood, the activity of GSH peroxidase, a selenoenzyme, was decreased by Hg. In liver, substantial accumulation of Hg with Se occured (molar Hg/Se ratio approximately 1.4 to 1.8) but GSH peroxidase activity was unaffected. We suggest that the coaccumulation of Hg and Se in liver measures the extent to which MeHg has been metabolically transformed by metabolism to Hg++, and inactivated by deposition as a Hg/Se complex of low bioavailability. The accumulation of Hg and Se in liver was much greater in cats fed tuna compared to pike, out of proportion to the relatively small differences in Hg and Se content of the tuna and pike basal diets. Some mechanisms are described by which selenium, vitamin E, and other factors might facilitate MeHg breakdown to inorganic Hg during long term low level exposure to MeHg.

Antimitogenic and proapoptotic activities of methylseleninic acid in vascular endothelial cells and associated effects on PI3K-AKT, ERK, JNK and p38 MAPK signaling.

Inhibiting the mitogenic response of vascular endothelial cells may in part mediate the antiangiogenic and anticancer activity of supranutritional selenium supplements. Our previous work had shown that methylseleninic acid (MSeA), a precursor of the critical anticancer methylselenol metabolite pool, was a potent inhibitor of the growth and survival of human umbilical vein endothelial cells (HUVECs). Here we investigated the effects of MSeA on selected protein kinase signaling transduction pathways to characterize their role in methylselenium induction of HUVEC cell cycle arrest and apoptosis. Exposure of asynchronous HUVECs for 30 h to 3-5 microM MSeA led to a profound G(1) arrest, and exposure to higher levels of MSeA not only led to G(1) arrest but also to DNA fragmentation and caspase-mediated cleavage of poly(ADP-ribose)polymerase, both biochemical hallmarks of apoptosis. Immunoblot analyses indicated that G(1) arrest induced by the sublethal doses of MSeA was associated with dose-dependent reductions of the levels of phospho-protein kinase B (also known as AKT or PKB), phospho-extracellular signal regulated kinase (ERK) 1/2, and phospho-Jun NH(2)-terminal kinases 1/2 in the absence of any change in p38 mitogen-activated protein kinase (MAPK) phosphorylation. Apoptosis induced by MSeA was associated with an increased phosphorylation of p38 MAPK in addition to the dephosphorylation of the above kinases. In HUVECs deprived of endothelial cell growth supplement (ECGS) for 48 h, resumption of ECGS stimulation resulted in an approximately 10-fold increase in mitogenic response, as indicated by [(3)H]thymidine incorporation into DNA. The ECGS-stimulated mitogenic response was inhibited in a dose-dependent manner by MSeA exposure with a IC(50) approximately 1 microM and a complete blockage at 3 microM. Wortmannin, an inhibitor of phosphatidylinositol 3-kinase (PI3K) upstream of AKT, potently inhibited the ECGS-stimulated DNA synthesis (IC(50), approximately 40 nM). Combining MSeA with Wortmannin showed an additive antimitogenic effect. An inhibitor of MAPK/ERK kinase 1, PD98059, also inhibited ECGS-stimulated DNA synthesis (IC(50), approximately 55 microM), but combining PD98059 with MSeA had an effect similar to that when PD98059 was used alone. A time-course experiment indicated that PI3K (AKT and ribosomal protein S6 kinase) activation occurred between 6 and 12 h of ECGS stimulation, and 3 microM MSeA exposure decreased AKT phosphorylation after 12 h of exposure, whereas no inhibitory effect was observed for ERK1/2 phosphorylation throughout the 30-h exposure duration. Additional experiments indicated that MSeA, Wortmannin, or a more specific PI3K inhibitor, LY294002, seemed to target, in the mid- to late-G(1) phase, a common mechanism(s) controlling G(1) progression to S while having no inhibitory effect on DNA synthesis once S-phase had initiated. Taken together, the results support a potent inhibitory activity at achievable serum levels of MSeA on ECGS-stimulated mitogenesis in the mid- to late-G(1) phase, and the target(s) of this inhibitory activity seems to be PI3K or components of this signal pathway. At pharmacological levels of exposure, modulation of ERK1/2 and other protein kinases may be relevant for the proapoptotic action of MSeA.

Thioredoxin reductase activity in rat liver is not affected by supranutritional levels of monomethylated selenium in vivo and is inhibited only by high levels of selenium in vitro.

Thioredoxin reductase is a selenoenzyme responsible for maintaining thioredoxin in the reduced form. Because thioredoxin is involved in many cellular processes, thioredoxin reductase is likely to be an important regulatory protein for both normal and transformed cells. Monomethylated selenium compounds inhibit carcinogenesis. In the present study, we investigated whether methylated forms of selenium would alter thioredoxin reductase activity in rats. The liver enzyme was used as a model system. Se-methylselenocysteine and methylseleninic acid consumed by rats at 2 microg Se/g diet for 3, 6, 10 or 22 wk did not affect activity compared with a basal diet containing 0.1 microg Se/g. The direct addition of 50 micromol dimethyl diselenide or dimethyl selenenylsulfide per L to liver extracts significantly inhibited thioredoxin reductase activity by approximately 60%. The magnitude of inhibition was dependent on the amount of thioredoxin in the assay and was reversible by dialysis, suggesting that a competitive type of inhibition occurs in vitro. Although thioredoxin reductase can be inhibited by high levels of selenium in a cell-free system, it should be noted that such a condition is unlikely to be attainable in vivo. Caution needs to be exercised in interpreting the in vitro results.

In vitro and in vivo studies of methylseleninic acid: evidence that a monomethylated selenium metabolite is critical for cancer chemoprevention.

Previous research suggested that the beta-lyase-mediated production of a monomethylated selenium metabolite from Se-methylselenocysteine is a key step in cancer chemoprevention by this agent. In an attempt to affirm the concept, the present study was designed to evaluate the activity of methylseleninic acid, a compound that represents a simplified version of Se-methylselenocysteine without the amino acid moiety, thereby obviating the need for beta-lyase action. The in vitro experiments showed that methylseleninic acid was more potent than Se-methylselenocysteine in inhibiting cell accumulation and inducing apoptosis in TM12 (wild-type p53) and TM2H (nonfunctional p53) mouse mammary hyperplastic epithelial cells, and these effects were not attributable to DNA damage, as determined by the comet assay. In general, methylseleninic acid produced a more robust response at one-tenth the concentration of Se-methylselenocysteine. It is possible that these cell lines may have only a modest ability to generate a monomethylated selenium species from Se-methylselenocysteine via the beta-lyase enzyme. In contrast, methylseleninic acid already serves as a preformed active monomethylated metabolite, and this could be an underlying reason why methylseleninic acid acts more rapidly and exerts a more powerful effect than Se-methylselenocysteine in vitro. Interestingly, the distinction between these two compounds disappeared in vivo, where their cancer chemopreventive efficacies were found to be very similar to each other [in both methylnitrosourea and dimethylbenz(a)anthracene rat mammary tumor models]. The beta-lyase enzyme is present in many tissues; thus, animals have an ample capacity to metabolize Se-methylselenocysteine systemically. Therefore, Se-methylselenocysteine would be expected to behave like methylseleninic acid if beta-lyase is no longer a limiting factor. Taken together, the present in vitro and in vivo results provide strong evidence in support of our earlier hypothesis that a monomethylated selenium metabolite is important for cancer chemoprevention. Methylseleninic acid could be an excellent tool, especially for molecular mechanism studies in cell culture, and some of these attributes are discussed.

Chemoprevention with triphenylselenonium chloride in selenium-deficient rats.

Cancer chemoprevention by high levels of selenium, including compounds like sodium selenite or selenomethionine, is generally not accompanied by increases in known selenoenzymes. There has been no information on whether selenoenzymes are obligatory mediators of the anticarcinogenic effect of selenium. Our previous experience with triphenylselenonium chloride suggests that it might be an ideal agent for studying selenium chemoprevention while simultaneously precluding the synthesis of selenoenzymes. Triphenylselenonium chloride has excellent tumor inhibitory activity but does not support the repletion of selenoenzymes in animals that have been deprived of a bioavailable form of selenium. In the present experiments, we evaluated the efficacy of mammary cancer protection by this compound in rats fed either a selenite-deficient (< 0.01 ppm Se) or selenite-adequate (0.1 ppm Se) diet. We also measured the activities of liver glutathione peroxidase and thioredoxin reductase as markers of selenium bioavailability in these different treatment conditions. In carcinogen-treated control animals not receiving triphenylselenonium chloride, mammary tumor incidence and the total number of tumors were similar between the selenite-deficient and selenite-adequate groups. Thus the correction of selenium deficiency by the addition of 0.1 ppm Se as selenite did not have detectable anticarcinogenic effects. Supplementation of triphenylselenonium chloride at a level of 30 ppm Se suppressed mammary tumorigenesis by approximately 50% regardless of dietary selenium nutritional status. However, this supplement had little effect on tissue selenium levels and did not increase liver glutathione peroxidase or thioredoxin reductase activities. In contrast, a level of 0.1 ppm Se as selenite did not affect mammary tumorigenesis but markedly increased tissue selenium concentrations and selenoenzyme activities. It is concluded that triphenylselenonium chloride does not release inorganic selenium for selenoprotein synthesis and that its anticancer activity involves mechanisms that are probably intrinsic to the compound. This study also shows for the first time that selenium chemoprevention is possible in an environment of severely depressed selenoenzyme expression. Thus selenium chemoprevention efficacy can be separated experimentally from selenoprotein synthesis using this model system.

Monomethyl selenium--specific inhibition of MMP-2 and VEGF expression: implications for angiogenic switch regulation.

Previous work suggested that antiangiogenic activity may be a novel mechanism contributing to the cancer chemopreventive activity of selenium (Se). Because methylselenol has been implicated as an in vivo active chemopreventive Se metabolite, experiments were conducted to test the hypothesis that this metabolite pool might inhibit the expression of matrix metalloproteinase-2 (MMP-2) by vascular endothelial cells and of vascular endothelial growth factor (VEGF) by cancer epithelial cells, two proteins critical for angiogenesis and its regulation. In human umbilical vein endothelial cells (HUVECs), zymographic analyses showed that short-term exposure to methylseleninic acid (MSeA) and methylselenocyanate (MSeCN), both immediate methylselenol precursors, decreased the MMP-2 gelatinolytic activity in a concentration-dependent manner. In contrast, Se forms that enter the hydrogen selenide pool lacked any inhibitory effect. The methyl Se inhibitory effect on MMP-2 was cell dependent because direct incubation with Se compounds in the test tube did not result in its inactivation. Immunoblot and enzyme-linked immunosorbent assay analyses showed that a decrease of the MMP-2 protein level largely accounted for the methyl Se-induced reduction of gelatinolytic activity. The effect of MSeA on MMP-2 expression occurred within 0.5 h of exposure and preceded MSeA-induced reduction of the phosphorylation level of mitogen-activated protein kinases (MAPKs) 1 and 2 (approximately 3 h) and endothelial apoptosis (approximately 25 h). In addition to these biochemical effects in monolayer culture, MSeA and MSeCN exposure decreased HUVEC viability and cell retraction in a three-dimensional context of capillary tubes formed on Matrigel, whereas comparable or higher concentrations of selenite failed to exert such effects. In human prostate cancer (DU145) and breast cancer (MCF-7 and MDA-MB-468) cell lines, exposure to MSeA but not to selenite led to a rapid and sustained decrease of cellular (lysate) and secreted (conditioned medium) VEGF protein levels irrespective of the serum level (serum-free medium vs. 10% fetal bovine serum) in which Se treatments were carried out. The concentration of MSeA required for suppressing VEGF expression was much lower than that needed for apoptosis induction. Taken together, the data support the hypothesis that the monomethyl Se pool is a proximal Se for inhibiting the expression of MMP-2 and VEGF and of angiogenesis. The data also indicate that the methyl Se-specific inhibitory effects on these proteins are rapid and primary actions, preceding or independent of inhibitory effects on mitogenic signaling at the level of MAPK1/2 and on cell growth and survival.

Selenium-induced inhibition of angiogenesis in mammary cancer at chemopreventive levels of intake.

The trace element nutrient selenium (Se) has been shown to possess cancer-preventive activity in both animal models and humans, but the mechanisms by which this occurs remain to be elucidated. Because angiogenesis is obligatory for the genesis and growth of solid cancers, we investigated, in the study presented here, the hypothesis that Se may exert its cancer-preventive activity, at least in part, by inhibiting cancer-associated angiogenesis. The effects of chemopreventive levels of Se on the intra-tumoral microvessel density and the expression of vascular endothelial growth factor in 1-methyl-1-nitrosourea-induced rat mammary carcinomas and on the proliferation and survival and matrix metalloproteinase activity of human umbilical vein endothelial cells in vitro were examined. Increased Se intake as Se-enriched garlic, sodium selenite, or Se-methylselenocysteine led to a significant reduction of intra-tumoral microvessel density in mammary carcinomas, irrespective of the manner by which Se was provided: continuous exposure (7-wk feeding) with a chemoprevention protocol or acute bolus exposure (3 d) after carcinomas had established. Compared with the untreated controls, significantly lower levels of vascular endothelial growth factor expression were observed in a sizeable proportion of the Se-treated carcinomas. In contrast to the mammary carcinomas, the microvessel density of the uninvolved mammary glands was not altered by Se treatment. In cell culture, direct exposure of human umbilical vein endothelial cells to Se induced cell death predominantly through apoptosis, decreased the gelatinolytic activities of matrix metalloproteinase-2, or both. These results indicate a potential for Se metabolites to inhibit key attributes (proliferation, survival, and matrix degradation) of endothelial cells critical for angiogenic sprouting. Therefore, inhibition of angiogenesis associated with cancer may be a novel mechanism for the anticancer activity of Se in vivo, and multiple mechanisms are probably involved in mediating the anti-angiogenic activity.

Selenium metabolism, selenoproteins and mechanisms of cancer prevention: complexities with thioredoxin reductase.

Numerous studies in animal models and more recent studies in humans have demonstrated cancer chemopreventive effects with Se. There is extensive evidence that monomethylated forms of Se are critical metabolites for chemopreventive effects of Se. Induction of apoptosis in transformed cells is an important chemopreventive mechanism. Apoptosis can be triggered by micromolar levels of monomethylated forms of Se independent of DNA damage and in cells having a null p53 phenotype. Cell cycle protein kinase cdk2 and protein kinase C are strongly inhibited by various forms of Se. Inhibitory mechanisms involving modification of cysteine residues in proteins by Se have been proposed that involve formation of Se adducts of the selenotrisulfide (S-Se-S) or selenenylsulfide (S-Se) type or catalysis of disulfide formation. Selenium may facilitate reactions of protein cysteine residues by the transient formation of more reactive S-Se intermediates. A novel chemopreventive mechanism is proposed involving Se catalysis of reversible cysteine/disulfide transformations that occur in a number of redox-regulated proteins, including transcription factors. A time-limited activation mechanism for such proteins, with deactivation facilitated by Se, would allow normalization of critical cellular processes in the early stages of transformation. There is uncertainty at the present time regarding the role of selenoproteins in chemoprevention model systems where supranutritional levels of Se are employed. Mammalian thioredoxin reductase is one selenoprotein that shows increased activity with Se supplementation in the nutritional to supranutritional range. Enhanced thioredoxin reduction could have beneficial effects in oxidative stress, but possible adverse effects are considered. Other functions of thioredoxin reductase may be relevant to cell signaling pathways. The functional status of the thioredoxin/thioredoxin reductase system during in vivo chemoprevention with Se has not been established. Some in vitro studies have shown inhibitory effects of Se on the thioredoxin system correlated with growth inhibition by Se. A potential inactivating mechanism for thioredoxin reductase or other selenoenzymes involving formation of a stable diselenide form resistant to reduction is discussed. New aspects of Se biochemistry and possible functions of new selenoproteins in chemoprevention are described.

Chemoprevention of mammary cancer with Se-allylselenocysteine and other selenoamino acids in the rat.

The present study examined the mammary cancer chemopreventive activity of Se-methylselenocysteine, Se-propylselenocysteine and Se-allylselenocysteine in the rat methylnitrosourea (MNU) model. Each compound was supplemented in the diet at a level of 2 ppm Se for the entire duration of the experiment after MNU dosing. Se-Allylselenocysteine was the most active and caused a reduction in total tumor yield by 86%. Se-Methylselenocyteine and Se-propylselenocysteine were similar but less effective, and both produced a decrease of about 50% in tumorigenesis. All three compounds were very well absorbed through the gastrointestinal tract. However, more selenium was excreted in urine after gavaging with Se-propylselenocysteine or Se-allylselenocysteine compared with Se-methylselenocysteine. Analysis of selenium in the mammary gland and other organs showed that tissue selenium levels did not appear to be correlated with differences in chemopreventive activity. A lyase activity capable of catalyzing scission of the Se-alkyl group from the remainder of the amino acid was demonstrated. This activity was found to be high in liver and kidney, but relatively low in mammary gland and intestine. Minimal variations in enzyme activity towards each of the substrates were observed. Our results support the concept that Se-alkylselenoamino acids could be used as precursors for delivering the Se-alkyl moiety and that intrinsic chemical differences in the Se-alkyl substituent of the test compounds are likely to be important determinants of their biological effects.

Activities of structurally-related lipophilic selenium compounds as cancer chemopreventive agents.

The present study compared the effects of four lipophilic forms of selenium with regard to cancer chemopreventive activity, tissue selenium accumulation, and bioavailability for synthesis of a selenoprotein. These reagents included methylphenyl selenide, diphenyl selenide, triphenyl-selenonium chloride, and p-xylylbis(methylselenide). The maximum tolerable dose (added in the diet) for each of these compounds was 5, 30, > 200, and 5 ppm Se, respectively. Because of differences in their tolerance, the cancer chemopreventive activities (in a methylnitrosourea-induced mammary tumor model in rats) of all 4 compounds were assessed at the 5 ppm Se level. Methylphenyl selenide was the most effective--79% inhibition, followed by p-xylylbis-(methylselenide)--66% inhibition, triphenylselenonium chloride--27% inhibition, and diphenyl selenide--10% inhibition. With respect to tissue selenium levels, p-xylylbis(methylselenide) produced the highest accumulation of selenium (approximately 3-fold increase in liver and kidney, 14-fold increase in mammary gland); methylphenyl selenide and diphenyl selenide showed more modest increases (1.5-fold or less in liver and kidney, 2.5-fold or less in mammary gland); while triphenylselenonium chloride resulted in no change. Highest bioavailability of selenium was observed for p-xylylbis(methylselenide), which was followed closely by methylphenyl selenide. Bioavailability was very low with diphenyl selenide, and undetectable with triphenylselenonium chloride. The chemical reactivities of these different selenium compounds are discussed in relation to the biological effects reported here.

Triphenylselenonium and diphenylselenide in cancer chemoprevention: comparative studies of anticarcinogenic efficacy, tissue selenium levels and excretion profile.

The objectives of the present study were to evaluate the cancer chemopreventive activity of triphenylselenonium chloride and diphenylselenide and to investigate the pharmacology of these two compounds with respect to their tissue accumulation and excretion profile. Although both phenyl selenide derivatives are related to each other structurally, they differ substantially in their intrinsic chemical properties. Triphenylselenonium is positively charged and amphiphilic, while diphenylselenide is uncharged and lipophilic. With the use of either the DMBA- or MNU-induced mammary tumor model in rats, triphenylselenonium was found to have superior chemopreventive efficacy compared to diphenylselenide. Both reagents were present at 30 ppm Se in the diet. At the time of sacrifice (22 weeks post-carcinogen), triphenylselenonium produced only minimal accumulation of selenium in the liver, kidney, mammary gland and plasma. In contrast, diphenylselenide caused a 2- to 3-fold elevation in selenium concentration depending on the tissue examined. Thus even though diphenylselenide was able to increase total selenium in tissues, it was less active in cancer protection. Fecal excretion following a single oral dose of triphenylselenonium (equal to the amount consumed in 1 day by an animal fed a diet containing 30 ppm Se) was approximately 78% and 8% of the dose during the first and second day, respectively, suggesting that the bulk of the dose was not absorbed. With diphenylselenide, fecal excretion was about 6% and 30% of the dose during the first and second day, and about 20% of the dose was excreted in the urine in each of the 2 days. This observation suggests that a large proportion of the diphenylselenide dose was absorbed and that urinary excretion was a major route of elimination for diphenylselenide once it was absorbed. Further studies are needed to clarify the basis for the differential effects of these phenyl selenide derivatives.

Effect on an aqueous extract of selenium-enriched garlic on in vitro markers and in vivo efficacy in cancer prevention.

Previous work has shown that the efficacy of cancer prevention by selenium-enriched garlic (Se-garlic) is primarily dependent on the action of selenium. An aqueous extract containing 43 micro Se/ml was prepared from lyophilized Se-garlic powder by the Soxhlet method. The activity of this Se-garlic extract was evaluated in a transformed mammary epithelial cell culture model for its effect on cell morphology, cell growth, cell cycle progression and the induction of single and double stranded breaks in DNA. Comparisons were also made with a similarly prepared extract from regular garlic, Se-methylselenocysteine (a major water-soluble seleno-amino acid identified in Se-garlic) and selenite (used for fertilizing Se-garlic). In contrast to the regular garlic extract which produced little or no modulation of the above parameters, treatment with the Se-garlic extract resulted in growth inhibition, GI phase cell cycle arrest and apoptotic DNA double strand breaks in the absence of DNA single strand breaks. This pattern of cellular responses was duplicated with exposure to Se-methylselenocysteine. Selenite, on the other hand, induced cell cycle blockage in the S/G2-M phase, and a marked increase in DNA single strand breaks (a measure of genotoxicity) in addition to growth suppression. The chemopreventive efficacy of the two garlic extracts was also investigated in the rat methylnitrosourea mammary tumor model. Both extracts were supplemented in the diet for 1 month immediately following carcinogen administration. Significant cancer protection was observed with treatment by the Se-garlic extract (at 3 p.p.m. Se in the diet), while little benefit was noted with treatment by the regular garlic extract. Based on the above in vitro and in vivo findings, it is hypothesized that the Se-garlic extract, in part via the action of Se-methylselenocysteine, is able to inhibit tumorigenesis by suppressing the proliferation and reducing the survival of the early transformed cells. Furthermore, the data also support the concept that the modulation of certain in vitro markers may be of value in predicting the effectiveness of novel forms of selenium for cancer prevention.

Dissociation of the genotoxic and growth inhibitory effects of selenium.

The effects of forms of selenium compounds that enter the cellular selenium metabolic pathway at different points were investigated in a mouse mammary carcinoma cell line. The goal of these experiments was to determine if the genotoxicity of selenium, defined as its ability to induce DNA single-strand breaks, could be dissociated from activities proposed to account for its cancer inhibitory activity. The results demonstrated that growth inhibition, measured as inhibition of cell proliferation and induction of cell death, was induced by all the forms of selenium evaluated. However, sodium selenite and sodium selenide, which are metabolized predominantly to hydrogen selenide, caused the rapid induction of DNA single-strand breaks as an early event that preceded growth inhibition. Interestingly methylselenocyanate and Se-methylselenocysteine, which are initially metabolized predominantly to methylselenol, induced growth inhibition in the absence of DNA single-strand breakage. Differences in the time course of selenium retention, in the occurrence of membrane damage, and in the induction of morphological changes by selenite versus methylselenocyanate were noted. Collectively, these data indicate that different pathways affecting cell proliferation and cell death are induced depending on whether selenium undergoes metabolism predominantly to hydrogen selenide or to methylselenol.

Activity of triphenylselenonium chloride in mammary cancer prevention.

The present study was designed to evaluate the tolerance and cancer chemopreventive activity of triphenylselenonium chloride in female Sprague-Dawley rats. No information is available in the literature on the anticarcinogenic efficacy of a lipophilic cationic selenium compound as exemplified by the triphenylselenonium ion. A short-term preliminary study indicated that it was well tolerated via the dietary route. Supplementation at levels up to 200 p.p.m. Se did not produce any apparent adverse effect in the animals. In the dimethylbenzanthracene mammary cancer model, a level of 30 p.p.m. Se in the diet reduced the total tumor yield by approximately 70% when treatment was applied during either the initiation phase or the post-initiation phase. In the MNU mammary cancer model, the inhibitory response was expressed only during the post-initiation phase. These findings suggest that the triphenylselenonium ion may have multiple modes of action in suppressing the development of neoplasia. Tissue analysis confirmed that there was minimal accumulation of total selenium until the level of supplementation reached 100 p.p.m. Se or above. Our study therefore convincingly demonstrates that triphenylselenonium chloride fits the criteria of an effective and desirable anticancer agent with a distinct separation between the chemopreventive dose range and the toxic dose range.

Selenium deficiency and decreased coenzyme Q levels.

The effect of long-term (18 months) selenium deficiency on the levels of liver coenzyme Q was studied in the rat. Levels of coenzyme Q9 and coenzyme Q10 in the liver of selenium-deficient rats were 40 and 67% of the levels in selenium-adequate animals, respectively. The results are similar to the findings using a shorter feeding period.

Significance of selenium-labeled proteins for selenium's chemopreventive functions.

A 58 kDa selenium-labeled protein purified from mouse mammary epithelial cells (MMEC) was used to examine whether selenium modulates protein synthesis or is just a marker for cellular selenium status. The protein was isolated using Sephadex G150 gel filtration and DEAE-Sephadex A50 ion-exchange chromatography. It was further analysed using 2-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and was found as a single spot with a pI of 4.6. The immunoreactivity with anti-58 kDa antiserum and the 75Se signal co-localized on a single 58 kDa protein band on both 1D- and 2D-PAGE. Partial amino acid analysis of the peptide showed homology with the thiol protein disulfide oxidoreductase (TPDO). Varying the selenium concentration in culture medium did not affect the protein content or the immunoreactivity of the 58 kDa protein. Additionally, selenium did not seem to regulate the activity of TPDO in TM6 cells. The glutathione peroxidase activity of TM6 cells, taken as the internal positive control, was enhanced with the increase in selenium concentration in the medium. The results suggest that selenium is attached to the 58 kDa protein, but does not regulate either its protein synthesis or its functional activity. We conclude that selenium labeling of the 58 kDa protein reflects the cellular selenium status but probably is not involved in its chemopreventive ability.

Selenium requirements of rats for normal hepatic and thyroidal 5'-deiodinase (type I) activities.

The nutritional requirement of selenium for type I 5'-deiodinase activity in thyroid compared with liver was assessed in rats. Male weanling Sprague-Dawley rats were fed a torula yeast-based diet for 20 wk. One group of rats was fed the Se-deficient basal diet (0.01 mg Se/kg). The other three groups were fed the basal diet plus sodium selenite at 0.05, 0.1 and 0.5 mg Se/kg diet. Liver 5'-deiodinase and glutathione peroxidase (GSH-Px) activities were depressed in the group fed the Se-deficient (basal) diet compared with the other groups. Liver 5'-deiodinase activity in the group fed 0.05 mg Se/kg diet was as high as in the groups fed 0.1 and 0.5 mg Se/kg diet, whereas GSH-Px activities in the groups fed 0.05 and 0.1 mg Se/kg diet were intermediate in value. Feeding the Se-deficient diet for 20 wk did not cause a suppression in 5'-deiodinase in the thyroid, and thyroid GSH-Px activity was approximately 40% of that in the other groups. In rats fed Se-supplemented diets, thyroid GSH-Px was approximately 20% or less of the activity found in liver. Plasma thyroxine was higher in the group fed the Se-deficient (basal) diet, but there were no differences in plasma 3,3',5-triiodothyronine among all groups. The results suggest that the nutritional Se requirement for 5'-deiodinase is less than that for GSH-Px and is approximately 0.05 mg Se/kg in the diet for normal activity in the liver and approximately 0.01 mg Se/kg for normal activity in the thyroid. Thyroid seems to be a priority organ over liver for Se when the intake of the element is limited.

Homocysteine-dependent demethylation of trimethylselenonium ion and selenobetaine with methionine formation.

In the presence of rat liver cytosol and homocysteine, trimethylselenonium ion (TMSe+) underwent time-dependent demethylation to dimethylselenide with the concurrent formation of methionine. Convenient methods were developed for assay of this activity using either radioactive methods based on the gamma emitting isotope 75Se or nonradioactive HPLC assay of methionine. The rate of demethylation was linear with protein concentration and dependent on homocysteine, which could not be replaced by cysteine, glutathione, or dithiothreitol. The TMSe+ demethylation rate was inhibited by the addition of betaine, sulfobetaine (dimethylthetin), or dimethylglycine. The Km for TMSe+ was 8 mM compared to 0.04 mM for betaine, but the rate of TMSe+ demethylation was approximately 50-fold that of betaine when both were assayed at 25 mM. Methionine was also produced from selenobetaine, selenobetaine methylester, and sulfobetaine. The selenium analogues of betaine inhibited the demethylation of TMSe+ with only minor decreases in methionine production, indicating substrate competition. In preliminary studies aimed at the partial purification of the TMSe+:homocysteine methyltransferase activity, the enzyme was found to have chromatographic and heat stability characteristics similar to betaine:homocysteine methyltransferase. The data indicate that betaine:homocysteine methyltransferase, or a very similar enzyme, is involved in the demethylation of TMSe+ and show that TMSe+, an in vivo urinary selenium metabolite of many selenium compounds, is not biologically inert.

Decreased ubiquinone levels in tissues of rats deficient in selenium.

The effect of selenium deficiency (-Se) on the levels of ubiquinones in liver, heart, kidney, and leg muscle was studied in the rat. Levels of ubiquinone 9 and ubiquinone 10 in the liver of -Se rats were about 50% of the levels in selenium adequate animals. Both ubiquinones in the heart were about 15% lower in -Se rats. Only ubiquinone 9 was significantly lower in the kidney of -Se rats. There was no difference in ubiquinone levels in leg muscle. Glutathione peroxidase activity in the tissues of -Se rats was > 95% lower. It is concluded that Se, as an integral part of the enzyme glutathione peroxidase, may protect tissues from oxidative damage, thereby preserving the ability of the cells to synthesize ubiquinone and preventing ubiquinone from oxidative degradation.