Search

Selenium has cancer protective effects in a variety of experimental systems. Currently, it is not known whether selenoproteins or low molecular weight selenocompounds are responsible for this activity. To evaluate the contribution of selenoproteins to the cancer protective effects of selenium, we used transgenic mice that carry a mutant selenocysteine transfer RNA gene, which causes reduced selenoprotein synthesis. Selenium homeostasis was characterized in liver and colon of wild-type and transgenic mice fed selenium-deficient diets supplemented with 0, 0.1, or 2.0 [micro]g selenium (as selenite)/g diet. [sup.75]Se-labeling, Western blot analysis, and enzymatic activities revealed that transgenic mice have reduced (P < 0.05) liver and colon glutathione peroxidase expression, but conserved thioredoxin reductase expression compared with wild-type mice, regardless of selenium status. Transgenic mice had more (P < 0.05) selenium in the nonprotein fraction of the liver and colon than wild-type mice, indicating a greater amount of low molecular weight selenocompounds. Compared with wild-type mice, transgenic mice had more (P < 0.05) azoxymethane-induced aberrant crypt formation (a preneoplastic lesion for colon cancer). Supplemental selenium decreased (P < 0.05) the number of aberrant crypts and aberrant crypt foci in both wild-type and transgenic mice. These results provide evidence that a lack of selenoprotein activity increases colon cancer susceptibility. Furthermore, low molecular weight selenocompounds reduced preneoplastic lesions independent of the selenoprotein genotype. These results are, to our knowledge, the first to provide evidence that both selenoproteins and low molecular weight selenocompounds are important for the cancer-protective effects of selenium. KEY WORDS: * cancer * colon * selenium * selenoproteins * transgenic mice

Evidence continues to implicate dietary components and genetic susceptibilities as important determinants of cancer risk and tumor behavior. Variation in cancer incidence among and within populations with similar dietary patterns suggests that an individual's response may reflect interactions with genetic factors, which may modify gene, protein, and metabolite expression patterns. Nutrigenomics, defined as the interaction between nutrition and an individual's genome, will likely provide important clues about responders and nonresponders. In this symposium, the role of bioactive food components in colon cancer susceptibility was used to exemplify the application of 'omic' technologies for cancer prevention. Topics that were addressed included dietary changes and gene polymorphisms (nutrigenetics), DNA methylation (nutritional epigenomics), gene expression (nutritional transcriptomics), altered formation or bioactivation of proteins (proteomics), and characterizing how the quantity and timing of exposure influence small molecular weight cellular constituents (metabolomics). The final presentation focused on exfoliated cells as a surrogate sample for the evaluation of bioactive food components in cancer prevention. The goal of the symposium was to provide an example of each of the 'omic' technologies as they relate to nutrition, cancer risk, and tumor behavior, and to help the participants understand that an integrated framework that simultaneously examines all of the 'omic' technologies is needed. KEY WORDS: * nutrigenomics * nutrigenetics * epigenetics * proteomics * metabolomics

Because manganese (Mn) is potentially toxic, and because dietary fat type may affect Mn absorption, the objectives of the current study were to determine whether diets containing very low or very high amounts of Mn and enriched in either saturated or unsaturated fats affected measures of neuropsychological and basic metabolic function. Healthy young women were fed for 8 wk each, in a crossover design, diets that provided 0.8 or 20 mg of Mn/d. One half of the subjects received 15% of energy as cocoa butter, and one half received 15% of energy as corn oil. A meal containing [sup.54]Mn was fed after 4 wk, and subjects underwent whole-body counting for the next 21 d. Blood draws and neuropsychological tests were administered at regular intervals during the dietary periods. When subjects consumed the diets low in Mn, compared with the high Mn diets, they absorbed a significantly higher percentage of [sup.54]Mn, but had a significantly longer biological half-life of the absorbed [sup.54]Mn. Manganese intake did not affect any neurological measures and only minimally affected psychologic variables. These data show that efficient mechanisms operate to maintain Mn homeostasis over the range of intakes that may be encountered in a mixed Western diet. Thus, dietary intakes of Mn from 0.8 to 20 mg for 8 wk likely do not result in Mn deficiency or toxicity signs in healthy adults. KEY WORDS: * manganese * fat * humans * toxicity * neurological

Several observations suggest a role for DNA methylation in cancer pathogenesis. Although both selenium and folate deficiency have been shown to cause global DNA hypomethylation and increased cancer susceptibility, the nutrients have different effects on one-carbon metabolism. Thus, the purpose of this study was to investigate the interactive effects of dietary selenium and folate. Weanling, Fischer-344 rats (n = 23/diet) were fed diets containing 0 or 2.0 mg selenium (as selenite)/kg and 0 or 2.0 mg folate/kg in a 2 x 2 factorial design. After 3 and 4 wk of a 12-wk experiment, 19 rats/diet were injected intraperitoneally with dimethylhydrazine (DMH, 25 mg/kg) and 4 rats/diet were administered saline. Selenium deficiency decreased (P < 0.05) colonic DNA methylation and the activities of liver DNA methyltransferase and betaine homocysteine methyltransferase and increased plasma glutathione concentrations. Folate deficiency increased (P < 0.05) the number of aberrant crypts per aberrant crypt foci, the concentration of colonic S-adenosylhomocysteine and the activity of liver cystathionine synthase. Selenium and folate interacted (P < 0.0001) to influence one-carbon metabolism and cancer susceptibility such that the number of aberrant crypts and the concentrations of plasma homocysteine and liver S-adenosylhomocysteine were the highest and the concentrations of plasma folate and liver S-adenosylmethionine and the activity of liver methionine synthase were the lowest in rats fed folate-deficient diets and supplemental selenium. These results suggest that selenium deprivation ameliorates some of the effects of folate deficiency, probably by shunting the buildup of homocysteine (as a result of folate deficiency) to glutathione. KEY WORDS: * selenium * folate * rats * DNA methylation * one-carbon metabolism

Cancer is a disease in which the cell cycle is altered, and the elucidation of the mechanisms by which constituents of human fecal water influence the cell cycle can lead to noninvasive measurement of colon cancer risk. The purpose of the present study was to investigate the effect of human fecal water on HT-29 cell-cycle progression with sodium selenite as a reference for comparison. Both human fecal water (2.5-5.0%) and selenite (3-4 [micro]mol/L) significantly inhibited cell growth. Cell-cycle analysis revealed that human fecal water decreased the proportion of S + G2 phase cells and increased that of G1 phase cells. In contrast, selenite decreased G1 phase cells and increased proportions of S and G2 phase cells. Both 5% human fecal water and 4 [micro]mol/L selenite greatly increased the mRNA level of the cyclin-dependent kinase inhibitor gene [p21.sup.waf1]. Interestingly, the mRNA levels of cyclin A and proliferating cell nuclear antigen (PCNA) were dramatically decreased by 69 and 62%, respectively, in HT-29 cells treated with fecal water but not selenite. In contrast, the mRNA level of DNA damage-inducible transcript 1, gadd45, was significantly increased by 2.28-fold in HT-29 cells treated with selenite but not fecal water. Furthermore, a PCNA gene promoter was cloned into a luciferase reporter construct and its activity was significantly reduced in a dose-dependent manner in cells treated with fecal water but not selenite. Collectively, these results suggest that human fecal water and selenite can differentially induce growth arrest genes, and that PCNA gene expression is uniquely and highly sensitive to human fecal water. KEY WORDS: * human fecal water * selenium * cell cycle * colon cell

A host of bioactive food components have been proposed to promote health and reduce the risk of disease states. It is clear that not all individuals respond identically to these essential and nonessential food components. Genetic polymorphisms may influence absorption, metabolism and accumulation of bioactive food components, thereby influencing their actions in target tissues. Unfortunately, serum concentrations of bioactive food components may not correlate with tissue concentrations and may therefore under- or overestimate the response in target tissues. Exfoliated cells may be useful to assess the actions of nutrients in specific tissues. Although not extensively examined, evidence already suggests the usefulness of these cells in predicting changes in gene expression, DNA methylation, DNA damage, protein expression and accumulation of dietary components. Although there are limitations on the collection of exfoliated cells, the inaccessibility of tissues they can represent raises intriguing possibilities for their ability to predict the outcome of nutritional intervention studies. J. Nutr. 133: 1769-1772, 2003. KEY WORDS: * exfoliated cells * biomarkers * cancer * colon * lung * breast

One possible dietary factor that may increase susceptibility to colon cancer is inadequate copper intake. The objective of this study was to investigate the effects of low and adequate copper intakes on copper nutriture and putative risk factors for colon cancer susceptibility in healthy men. Seventeen healthy free-living nonsmoking men aged 21-52 y completed a 13-wk controlled feeding study in a randomized crossover design. The basal diet contained 0.59 mg Cu/13.65 MJ. After a 1-wk equilibration period in which the men consumed the basal diet supplemented with 1.0 mg Cu/d, they were randomly assigned to receive either the basal diet or the basal diet supplemented with 2 mg Cu/d for 6 wk. After the first dietary period, the men immediately began to consume the other level of Cu for the last 6 wk. They collected their feces during the equilibration period and during the last 2 wk of the two dietary periods for free radical and fecal water analysis. Low dietary copper significantly (P < 0.01) increased fecal free radical production and fecal water alkaline phosphatase activity. Low dietary copper significantly (P < 0.0001) decreased fecal water copper concentrations but did not affect fecal water volume, pH, iron or zinc concentrations. In contrast to the fecal analysis, hematological indicators of copper status were not significantly affected by the dietary treatments. These results suggest that low dietary copper adversely affects fecal free radical production and fecal water alkaline phosphatase activity, which are putative risk factors for colon cancer. KEY WORDS: * copper * fecal water * free radical production * alkaline phosphatase activity * cytotoxicity

Previous studies have show that changes in protein kinase C (PKC) isoform expression may be related to increased susceptibility of copper-deficient rats to aberrant crypt formation. The purpose of this study was to determine whether dietary copper would affect azoxymethane-induced intestinal tumor formation and PKC isozyme expression in normal colonic mucosa and tumor samples. Eighty weanling Fischer-344 rats were randomly assigned to diets that contained either 0.8 or 5.3 [micro]g Cu/g diet. After 24 and 31 d of diet consumption, 30 rats/diet were administered azoxymethane (15 mg/kg i.p.) and 10 rats/diet were administered saline. Rats continued to consume their respective diets for an additional 38 wk. Rats injected with azoxymethane and fed the low copper diet had a significantly (P < 0.0001) greater small intestinal and total tumor incidence compared with rats fed adequate dietary copper. However, dietary copper did not affect colon tumor incidence. Low dietary copper significantly (P < 0.004) decreased PKC [alpha] protein expression in normal but not in tumor tissue. In contrast, low dietary copper did not affect PKC [delta] or [zeta] protein expression in either the normal or tumor tissue. PKC [alpha] and [delta] protein and mRNA expression were lower in tumor tissue than in normal tissue. These results along with previous observations suggest that dietary copper-mediated changes in PKC [alpha], [delta] and [zeta] protein expression are not as important for colon tumor promotion/progression as they are for tumor initiation. J. Nutr. 132: 1018-1025, 2002. KEY WORDS: * colon cancer * protein kinase C * copper * rats

Several observations implicate a role for altered DNA methylation in cancer pathogenesis. The global level of DNA methylation is generally lower; however, DNA methyltransferase (Dnmt1) activity is usually higher in tumor cells than in normal cells. The purpose of this study was to investigate whether the Dnmt1 inhibitor, 5-aza-2'-deoxycytidine (aza-dC) would alter the effect of dietary selenium on the formation of aberrant crypts. Weanling rats (n = 60) were fed three concentrations of selenium (deficient, 0.1 and 2.0 mg/kg diet) in a Torula yeast-based diet. Half of the rats were injected weekly with aza-dC (1 mg/kg, subcutaneously) and half were injected with the vehicle control (PBS). After 3.5 wk of consuming the experimental diets, the rats were given two injections of dimethylhydrazine (DMH; 25 mg/kg, intraperitoneally). Rats fed the selenium-deficient diet and injected with PBS had significantly (P < 0.006) more aberrant crypts than rats fed 0.1 or 2.0 mg selenium/kg diet (244 [+ or -] 21 vs. 165 [+ or -] 9 and 132 [+ or -] 14, respectively). In contrast, when rats were injected with aza-dC, there was a significant (P < 0.0001) reduction in aberrant crypt formation and dietary selenium had no effect (62 [+ or -] 8 vs. 77 [+ or -] 13 vs. 54 [+ or -] 8, in rats fed 0, 0.1 and 2.0 mg selenium/kg diet, respectively). HT-29 cells cultured in the absence of selenium had significantly hypomethylated DNA but significantly more Dnmt1 protein expression than cells cultured in the presence of 1 or 2 [micro]mol/L selenium. These results suggest that aza-dC treatment may protect selenium-deficient rats against carcinogen-induced aberrant crypt formation. J. Nutr. 132: 292-297, 2002. KEY WORDS: * selenium * DNA methylation, azadeoxycytidine * HT-29. DNA methyltransferase

Aberrant crypt loci (ACF) are preneoplastic lesions for colon cancer. Altered amounts of copper-zinc (CuZnSOD) and manganese (MnSOD) superoxide dismutases have been implicated in multistage carcinogesis of both rodents and humans. Dietary factors are potential modulators of both CuZnSOD and MnSOD activity. The purpose of this study was to investigate the interactive effects of dietary copper, manganese, and iron on 3,2[prime]-dimethyl-4-aminobiphenyl (DMABP)-induced ACF and superoxide dismutase activities in weanling rats fed low or adequate copper (0.8 or 5.1 [[micro]gram] Cu/g diet), low or adequate manganese (0.6 or 17 [[micro]gram] Mn/g diet), and adequate or high iron (37 or 140 [[micro]gram] Fe/g diet). Twelve rats were allowed free access to each of these eight diets for 3.5 wk prior to DMABP administration and for an additional 8 wk after the first DMABP injection. Rats fed low dietary copper had 105% (P < 0.0001) higher formation of DMABP-induced ACF than those fed adequate dietary copper. Rats ingesting low rather than adequate dietary manganese had 23% higher formation of ACF, and rats ingesting high rather than adequate dietary iron had 18% higher formation of ACF. Heart total superoxide dismutase activity was significantly correlated with the number of ACF (r = -0.43, P < 0.0001) in rats administered DMABP. These results suggest that dietary alterations that affect superoxide dismutase activity may affect cancer susceptibility. KEY WORDS: copper; manganese; iron; aberrant crypt loci; superoxide dismutase; rats

There is increasing evidence that selenium can protect against tumorigenesis or preneoplastic lesion development induced by chemical carcinogens. This study examined whether selenite, selenate or selenomethionine would be protective against 3,2[prime]-dimethyl-4-aminobiphenyl (DMABP)-DNA adduct formation in the liver and colon of rats and sought to delineate the mechanism for the protective effects of the different chemical forms of selenium against aberrant crypt formation, a preneoplastic lesion for colon cancer. After injection of DMABP, two DNA adducts were identified in the liver and colon of rats. Supplementation with either 0.1 or 2.0 mg selenium/kg diet as either selenite or selenate but not selenomethionine resulted in significantly fewer (53-70%; P < 0.05) N-(deoxyguanosin-8-yl)-(deoxyguanosin-8-yl)-3,2[prime]-dimethyl-4-aminobiphenyl (C8-DMABP)-DNA adducts in the colon but not the liver than in rats fed a selenium-deficient diet. Rats supplemented with selenomethionine had greater (P < 0.05) plasma and liver selenium concentrations and glutathione peroxidase activity than those supplemented with selenite or selenate; however, they also had more DMABP-DNA adducts. The protective effect of selenite and selenate against DMABP-DNA adduct formation apparently is not a result of alterations in plasma or liver selenium concentrations or altered glutathione peroxidase or glutathione transferase activities but may be related to differences in the metabolism of the different forms of selenium. KEY WORDS: selenium; rats; DNA adducts; colon cancer

Dietary selenium intake is reported to be inversely associated with colon cancer risk in several epidemiological and preclinical studies. To determine the importance of selenoproteins in cancer protection, we conduced studies to characterize selenium homeostasis in transgenic mice that express a mutant Sec tRNA[[Ser].sup.Sec] lacking the isopentenyladenosine ([i.sup.6]A) base at position 37 and have selectively reduced selenoprotein synthesis. Wild-type (wt) and [i.sup.6]A-deficient mice were fed diets containing 0, 0.1, or 2.0 [micro]g selenium (as selenite)/g diet (deficient, adequate, and supplemental selenium, respectively; n = 5-6 mice/group). Compared with wt mice, liver and colon tissue selenium concentrations were lower in [i.sup.6]A-deficient mice fed adequate selenium (1014 [+ or -] 67 vs. 252 [+ or -] 14 and 225 [+ or -] 11 vs. 63 [+ or -] 3 ng/g, respectively; P < 0.001). In contrast, when mice consumed supplemental selenium, there were no significant differences in liver selenium concentrations between the two strains. Western blot and enzymatic analysis revealed that [i.sup.6]A-deficient mice had significantly reduced (P < 0.001) glutathione peroxidase 1 (GPX1) expression and activity in both liver and colon tissue, which was not corrected by supplemental selenium. However, neither mouse strain nor dietary selenium significantly affected protein levels of thioredoxin reductase (TR1), selenoprotein P, a 15-kDa selenoprotein, and GPX2. Compared with those fed the selenium-deficient diet, both wt and [i.sup.6]A-deficient mice fed adequate and supplemental selenium diets had significantly increased (P < 0.05) TR1 activity in liver and colon tissue, [i.sup.6]A-Deficient mice had significantly more (P < 0.005) azoxymethane-induced aberrant crypts and aberrant crypt loci than wt mice. Compared with adequate selenium, supplemental selenium decreased aberrant crypt formation in both wt and [i.sup.6]A-deficient mice. In summary, supplemental selenium can increase tissue selenium concentrations and TR activity in [i.sup.6]A-deficient mice but does not rescue GPX1 expression or activity. These results also suggest that selenoproteins are important for the cancer-protective effects of selenium.