Environmental and Molecular Mutagenesis 00:000^000 (2009) Research Articles Increased Consumption of Wheat Biofortified With Selenium Does Not Modify Biomarkers of Cancer Risk, Oxidative Stress, or Immune Function in Healthy Australian Males Jing Wu,1,2 Carolyn Salisbury,1 Robin Graham,2 Graham Lyons,2 and Michael Fenech1* 1 Nutritional Genomics and Genome Damage Diagnostics Laboratory, CSIRO Human Nutrition, Food Science Australia, Adelaide, South Australia, Australia 2 Discipline of Plant and Food Science, School of Agriculture, Food and Wine, University of Adelaide, Glen Osmond, South Australia, Australia Increased intake of selenium (Se) may reduce the risk of degenerative diseases including cancer but excessive intake may be toxic. Wheat is a major source of dietary Se in humans. However, the effect of Se from wheat that is agronomically biofortified with Se on biomarkers of human health status is unknown. This study aimed to investigate whether improving Se status, by increased dietary intake of Se-biofortified wheat, affects biomarkers of cancer risk, cardiovascular disease risk, oxidative stress, and immune function in healthy South Australian men. A 24-week placebo-controlled double-blind intervention was performed in healthy older men (n 5 62), with increased dose of Se intake every 8 weeks. Wheat was provided as 1, 2, and 3 puffed wheat biscuits, during weeks 1–8, 9–16, and 17–24, respectively. Blood was collected to measure a wide range of disease risk biomarkers. Consumption of Se-biofortified wheat was found to increase plasma Se concentration from a baseline level of 122 to 192 lg/L following intake of three biscuits/day, which provided 267 lg Se. Platelet glutathione peroxidase, chromosome aberrations, and DNA damage in lymphocytes measured using the cytokinesis-block micronucleus cytome assay and with the Comet assay, plasma F2-isoprostanes, protein carbonyls, plasma C-reactive protein, and leukocyte number were unaffected by the improved Se status. Improvement of Se status by consumption of Se-biofortified wheat did not substantially modify the selected biomarkers of degenerative disease risk and health status in this apparently selenium-replete cohort of healthy older men in South Australia. Environ. Mol. Mutagen. 00:000–000, 2009. VC 2009 Wiley-Liss, Inc. Key words: selenium; biofortified wheat; chromosome damage; DNA damage; oxidative stress; immune function INTRODUCTION Moderate deficiency of selenium (Se) is associated with various pathological conditions including infertility, increased oxidative stress, immune dysfunction, cognitive impairment, and increased risk for specific cancers such as prostate cancer [Rayman, 2000, 2005; Gronberg, 2003; Bjelakovic et al., 2004; Gromadzinska et al., 2008]. Several essential structural proteins and enzymes in the body have improved function when seleno-amino acids such as seleno-methionine (Se-met) and seleno-cysteine are incorporated into the protein instead of their sulphur-containing amino acid analogs [Gromer et al., 2005; Hatfield et al., 2006]. There is emerging evidence that increased intake of organic Se may reduce the risk of certain degenerative diseases, but there is also concern that excessive Se C 2009 Wiley-Liss, Inc. V intake may have unwanted toxic effects, such as selenosis [Yang and Zhou, 1994; Reid et al., 2004]. Within the physiological concentration range of 3 to 430 lg Se/L, Grant sponsors: National Center of Excellence for Functional Foods (via the National Food Industry Strategy), HarvestPlus/The University of Adelaide, The South Australian Grains Industry Trust Fund, Laucke Flour Mills. *Correspondence to: M. Fenech, CSIRO Human Nutrition, PO Box 10041 Gouger Street, Adelaide BC, SA 5000, Australia. E-mail: michael.fenech@csiro.au Received 29 October 2008; provisionally accepted 12 February 2009; and in final form 13 February 2009 DOI 10.1002/em.20490 Published online in Wiley InterScience (www.interscience.wiley.com). Environmental and Molecular Mutagenesis. DOI 10.1002/em 2 Wu et al. Se, as Se-met, had no impact on baseline micronucleus frequency or g-ray induced chromosome damage in human lymphocytes in vitro; however, spontaneous frequencies of nucleoplasmic bridges and nuclear buds declined significantly as the dose increased but higher concentrations of Se-met caused strong inhibition of nuclear division and increased cytotoxicity [Wu et al., 2009]. Dietary selenium supplementation in dogs reduced DNA damage in prostate tissue as measured by the alkaline Comet assay but was not associated with glutathione peroxidase (GPx) activity in plasma; however, excessive intake of Se appeared to increase DNA damage suggesting a U-shaped dose-response [Waters et al., 2003, 2005]. In a study of men at high risk for prostate cancer, DNA damage in lymphocytes measured by Comet assay was shown to be inversely associated with serum Se concentration for those with serum Se less than 98 lg/L but not for those with higher concentrations [Karunasinghe et al., 2004]. Increased Se intake has been associated with decreased risk for cardiovascular disease (CVD) but it is unknown whether this is due to Se-mediated reduction in lipid peroxidation, inhibition of inflammation, or an improved lipid profile in the blood [Ravn-Haren et al., 2008]. There is also a need to know that both CVD and cancer risk biomarkers are affected favorably and thus verify that benefits, if present, occur across multiple conditions of degenerative disease. In countries where wheat products are widely consumed, wheat is usually a major source of dietary Se. In Australia, for example, it is estimated that most people obtain around half their Se from wheat [Lyons et al., 2004, 2005a]. The Se content of wheat can be increased by agronomic biofortification. This involves fertilizing the growing crop with an appropriate inorganic form of the micronutrient, which the plant converts to several organic Se forms, notably Se-met, which are more suitable for human consumption [Lyons et al., 2003, 2005b]. The effect of increased consumption of Se via Se-biofortified wheat on genome damage and immune function has not been tested previously. The hypothesis of this study was that improving Se status of older South Australian men by increased intake of Se-biofortified wheat has a beneficial impact on biomarkers of risk for cancer, oxidative stress, and immune function. To the best of our knowledge, the results of this study are the most comprehensive assessment of the bioefficacy and safety of Se-biofortified wheat performed so far. MATERIALS AND METHODS Wheat Used in the Intervention Trial Wheat (cultivar Whylah) was biofortified with Se in October 2003 by applying sodium selenate as a foliar spray at around flowering time on a farm near Frances, in the South-East region of South Australia. The biofortified grain was analyzed by a fluorimetric method [Watkinson, 1966; Koh and Benson, 1983] and found to contain 10 mg/kg Se. Control wheat (cultivar Yitpi) was low in Se (0.07 mg/kg Se) but similar to the biofortified wheat in protein and minerals (including Fe, Zn, Cu, Mn, S, Ca, Mg) (data not shown). Study Design The study design was a double-blind placebo-controlled intervention with a dose-response. The study was advertised in the local papers and in electronic media from September to November 2004. The intervention commenced in early February 2005 and was completed in early August 2005. Respondents were initially screened for eligibility using the following inclusion criteria: healthy males, aged 40–70 years, not supplementing with selenium and not supplementing with above recommended daily intake (RDI) levels of folate and/or vitamin B12 and/or vitamin C. The following respondents were excluded from the study: (a) cancer patients undergoing chemotherapy or radiotherapy, (b) those with sensitivity to study foods, i.e., gluten/wheat intolerance, (c) those unable to comprehend or comply with the study protocol, and (d) those not available for all sampling phases of the study. A total of 179 men, living in or near Adelaide, South Australia, were eligible for the study and were screened for plasma Se level in peripheral blood samples collected after an overnight fast. The 81 men with the lowest plasma Se concentration were then admitted into the trial and randomized to three dietary groups (N 5 27 per group) as shown in Figure 1. The study was focused on men with lower Se status because it was expected that any beneficial effects of improved Se status had a higher probability of being observed in this group and because Se status is of particular relevance to men given its association with reduced prostate cancer risk. The dietary groups were ‘‘CONTROL,’’ ‘‘BIOFORT,’’ and ‘‘PROFORT’’ depending on the wheat source of the biscuits they were required to consume. Trial participants were required to consume one biscuit per day for the first 8 weeks, then two biscuits daily for the next 8 weeks, then three biscuits daily for the final 8 weeks. The intention of the study design was that each BIOFORT and PROFORT biscuit would deliver 75 lg Se so that the daily amount of Se from the biscuits would increase progressively from 75 lg, to 150 lg, to 225 lg during each phase of the trial. The biscuits were developed by Laucke Flour Mills (Strathalbyn, South Australia) and were made by soaking whole grain wheat in water for 24 hr, then heating to expand the grain, and compressing into a ‘‘puffed wheat’’ biscuit. As well as the biofortified (BIOFORT) wheat biscuits and the low-Se control biscuits (CONTROL), a process-fortified (PROFORT) ‘‘positive control’’ biscuit was developed by adding pure Se-met (Eburon Organics, USA) to the water which was absorbed by the grains. The PROFORT control was included to compare the effect of Se provided by biofortification with that provided by process-fortification. The biscuits used in the trial were made in three separate batches before the trial commenced, their weight and Se concentration were monitored throughout the trial, and were found to remain constant during the trial period. The actual Se content per biscuit (mean value (range) in lg) was 0.71 (0.64–0.75), 89.1 (86.4–94.5), and 101.9 (97.0–105.8) for CONTROL, BIOFORT, and PROFORT biscuits, respectively. We had intended the Se concentration in PROFORT and BIOFORT biscuits to be identical however due to a combination of higher Se concentration and greater weight, the PROFORT biscuits contained 18% more total Se than the BIOFORT biscuits. The three types of biscuits were completely indistinguishable from each other. None of the staff directly involved in the trial, sample analyses or participants had any knowledge of the selenium level or fortification process of the various biscuit groups. The total estimated daily Se intakes of the trial participants from the provided biscuits in each group are presented in Figure 1. To assess compliance participants were required to keep a re- Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers 3 Fig. 1. Trial design. Asterisk indicates blood samples collected at the beginning of the indicated week. CONTROL, BIOFORT, AND PROFORT groups consumed biscuits made from normal wheat, wheat biofortified with selenium, and wheat process fortified with selenomethoinine, respectively. 1, 2, and 3 biscuits were consumed daily between weeks 0– 8, weeks 8–16, and weeks 16–24, respectively. The CONTROL, BIOFORT, and PROFORT biscuits contained 0.7, 89.1, and 101.9 lg Se each, respectively. Estimated Se intakes per day (lg/day) at the different stages of the trial are also indicated in the diagram. cord of the number of biscuits they had eaten and return any biscuits that were not consumed. Dietary intake of selenium before the intervention commenced was measured using a validated food frequency questionnaire [Baghurst and Record, 1984; Baghurst et al., 1992]. Fasted blood samples were collected at the start (wk 0) and after 8 (wk 8), 16 (wk 16), and 24 (wk 24) weeks of the trial. The study was approved by the Human Ethics Committee of Commonwealth Scientific and Industrial Research Organisation (CSIRO) Human Nutrition. nucleus assay may be increased in those with low blood concentration of folate and/or vitamin B12 [Fenech, 1998]. Main and Secondary Outcome Measures Main outcome measures were (i) plasma Se as a biomarker of bioavailability, (ii) platelet GPx as a biomarker of seleno-enzyme activity, and (iii) chromosome and DNA damage in lymphocytes measured using the cytokinesis-block micronucleus cytome (CBMN Cyt) assay and the Comet assay. The micronucleus frequency index in the CBMN Cyt assay has been shown to be associated prospectively with cancer risk and the micronucleus, nucleoplasmic bridge, and nuclear bud indices have been shown to be strongly associated with lung cancer risk in smokers [Bonassi et al., 2007; El-Zein et al., 2008]. Other secondary bioefficacy biomarkers included (a) plasma F2-isoprostanes and protein carbonyls as indicators of oxidative stress, (b) plasma C-reactive protein (CRP) as a biomarker of inflammation, and (c) leukocyte number and subsets as biomarkers of immune function. Plasma folate and vitamin B12 were also measured to assess whether there were substantial dietary changes during the trial affecting these parameters which may have impacted on the DNA damage markers. Previous studies showed that chromosome damage measured using the micro- Methods for Outcome Measures Se concentrations in wheat, biscuits, and plasma were determined by inductively coupled plasma mass spectrometry (ICPMS) (Agilent Technologies 7500c, Japan), following digestion with nitric acid and hydrochloric acid. Accuracy was assured by analysis of certified reference material (Seronorm Serum Lot JL4409). There was only 1.5% variation from the expected value of the Seronorm reference material. Se speciation studies to determine concentration of Se-met and methionine selenoxide were performed using protease digestion and isotope dilution HPLC coupled to ICP-MS [Kirby et al., 2006]. Platelet glutathione peroxidase activity was measured spectrophotometrically using the method of Misso et al. [1996] and a GPx assay kit (Sigma catalogue # CGP-1). Lymphocyte chromosome damage was measured using the CBMN Cyt assay as described by Fenech [2007] using whole blood cultures of fresh blood collected within 2 hr before commencing the assay. Frequency of binucleated cells with micronuclei (MN BNCs, a biomarker of chromosome breakage or loss), nucleoplasmic bridges (NPB BNCs, a biomarker of DNA misrepair or telomere end fusion), and nuclear buds (NBUD BNCs, a biomarker of gene amplification) was measured in 1,000 BNCs. Ratios of mono-, bi-, and multinucleated cells were measured to determine the nuclear division index (NDI, a biomarker of mitogen and immune responsiveness). The slides were scored manually by a single scorer (CS). Lymphocyte DNA damage by the Comet assay was measured using the alkaline method as described by Collins [2005] in both fresh and cry- Environmental and Molecular Mutagenesis. DOI 10.1002/em 4 Wu et al. TABLE I. P Values for Effect of Treatment and Time on Biomarkers Measured in the Trial Biomarkers Se status Selenoprotein activity DNA damage in lymphocytes Comet assay DNA damage in lymphocytes CBMN Cyt assay Immune function Leucocyte counts Oxidative stress and inflammation B vitamin status Plasma Se Platelet GPx activity Fresh lymphocytes Category 4 Fresh lymphocytes Category 0–3 Frozen lymphocytes Category 4 Frozen lymphocytes Category 0–3 MN BNCs NPB BNCs NBUD BNCs NDI CD4:CD8 ratio CD8 lymphocytes CD4 lymphocytes CD3 lymphocytes Neutrophils Lymphocytes Total white cells Protein carbonyls F2-isoprostanes C-creative protein Plasma B12 Plasma folate Effect of treatment (P value) Effect of time (P value) 0.006 0.715 0.007 0.074 0.719 0.0001 0.146 0.011 0.214 0.061 0.193 0.825 0.868 0.600 0.591 0.146 0.435 0.422 0.456 0.080 0.106 0.876 0.174 0.621 0.391 0.334 0.487 0.207 0.622 0.074 0.514 0.592 0.707 0.001 0.180 0.199 0.043 0.228 0.234 0.004 0.421 0.091 0.917 0.039 GPx, glutathione peroxidase; CBMN Cyt assay, cytokinesis-block micronucleus cytome assay; MN BNCs, binucleated cells with micronuclei; NPB BNCs, binucleated cells with nucleoplasmic bridge; NBUD BNCs, binucleated cells with nuclear bud; NDI, nuclear division index. P values in bold are statistically significant. opreserved isolated lymphocytes. We used both fresh and cryopreserved lymphocytes because in the latter case we could analyze the cells from each time-point and each individual within the same assay and thus minimize the effect of day-to-day assay variation on the results obtained. To avoid confounding by apoptotic cells we scored Category 4 cells (with most of the DNA in the comet tail) separately from those in Category 0– 3. The slides were scored manually by a single scorer (JW) as per criteria published elsewhere [Collins, 2002]. Plasma F2-isoprostanes were measured by gas chromatography mass spectrometry using the method of Mori et al. [1999]. Protein carbonyls in plasma were measured by ELISA using the method of Buss et al. [1997]. White blood cell counts, lymphocyte subsets, plasma CRP, plasma vitamin B12, and folate were measured by the Institute of Medical and Veterinary Sciences (Adelaide, South Australia) in their certified routine diagnostic laboratory. All plasma measures were performed on cryopreserved samples stored at 2808C after the intervention was completed. Statistical Analysis The sample size and study power estimates were based on the assumption that at least 20 subjects per group would complete the intervention with high compliance and designed to detect a change of (a) 5% in plasma Se concentration, (b) 32% change in the frequency of MN BNCs in the CBMN-Cyt assay with 90% power and P < 0.05, and (c) to detect a difference of 13 arbitrary units (AU) of DNA damage measured using the Comet assay in lymphocytes between two groups with 80% power and P < 0.05, two-tailed. These estimates were calculated using historical standard deviation values of 5.8 lg/L (plasma Se assay) and 2.6 MN BNCs per 1,000 BNCs (CBMN Cyt assay) for subjects within the age and gender group relevant to this study. For the Comet assay, the sample size required to detect a statistically significant difference was predetermined based on the published data on 41 male healthy nonsmokers aged 40–55 yrs, in which the mean 6 1 SD of DNA damage (arbitrary unit, AU) in lymphocytes by Comet assay was 82.3 6 14.1 [Piperakis et al., 2003]. QQ-plots on standard residues of all outcome measures were performed to test the normality of the data sets. Difference between distributions of smoking and alcohol status of three study groups were determined by Chi-square analysis. One-way ANOVA was used to compare the difference of baseline characteristics between treatment groups, such as age, BMI, plasma Se, as well as the changes of GPx value between three treatment groups following the interventions after the baseline GPx values were stratified into low, medium, and high tertile. The significance of effect of treatment and time for each parameter was measured using general linear model repeated measures mixed between-within subjects ANOVA [Pallant, 2005] on the delta value of each follow-up time point (i.e., baseline value subtracted from the actual value measured at that time-point) with baseline values included as covariates to take account of effect of the baseline value on the delta value. Cross correlation analysis of relationships between all biomarkers (at the start or the end of the trial) were conducted using partial correlation test, after controlling for any effect of age. Differences with P value <0.05 were considered to be statistically significant. Statistical analyses were performed using the statistical package SPSS for WINDOWS (version 16.0, SPSS Inc, Chicago). RESULTS Results for all biomarkers are summarised in Tables I–IX. 25, 24, and 24 participants started and 22, 19, and 21 com- Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers 5 TABLE II. Cross Correlations Between Biomarkers at Baseline Analyzed Using Partial Correlation After Adjusting for Age r Value NBUD Trig HDL LDL CRP WCC Lymp Neut CD3 CD4 CD8 Comet Comet CD4:8 (frozen) (fresh) Platelet Plasma GPx Se NDI NS NS NS NS NS NS NS NS NS NS NS NS NS 20.27* NS NS MN 0.42** NS NS NS NS NS NS NS NS NS NS NS NS 0.34** NS NS NPB 0.40** NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NBUD — NS NS NS NS NS NS NS NS NS NS NS NS 0.25* NS NS Protein NS NS NS NS NS 0.26* NS NS NS NS 0.27* NS 0.36** NS NS carbonyl Chol 0.28* NS 0.87** NS NS NS NS NS NS NS NS NS NS NS NS Trig — 20.47** NS NS 0.31* 0.35** NS NS 0.26* NS NS NS NS NS NS HDL — NS 20.25* 20.34** 20.48** NS NS NS NS NS NS NS NS NS CRP — 0.31* 0.36** NS NS NS NS NS NS NS NS 20.25* WCC — 0.68** 0.94** NS 0.41** NS 0.26* 0.32* NS NS NS Lymp — 0.40** 0.42** NS NS NS NS NS NS NS Neut — NS 0.43** 20.29* 0.37** 0.42** NS NS NS CD3 — 0.31* 0.51** 20.35** NS NS 20.35** 20.31* CD4 — 20.62** 0.66** NS NS 20.31* 20.27* CD8 — 20.88** NS 20.29* NS NS CD4:8 — NS 20.4** NS NS Comet — 0.29* NS NS (frozen) NS, nonsignificant, r values 2-tailed; NDI, nuclear division index; MN, binucleated cells with micronuclei; NPB, binucleated cells with nucleoplasmic bridge; NBUD, binucleated cells with nuclear bud; Chol, total plasma cholesterol; Trig, plasma triglyceride; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CRP, C-reactive protein; WCC, white cell count; Lymp, lymphocytes; Neut, neutrophils; CD3, CD3-positive lymphocytes; CD4, CD4-positive lymphocytes; CD8, CD8-positive lymphocytes; CD4:8, ratio of CD4-positive: CD8-positive lymphocytes; GPx, glutathione peroxidase; NS, nonsignificant. *P < 0.05; **P < 0.01. TABLE III. Correlation Between Biomarkers at Week 24 Analyzed Using Partial Correlation After Adjusting for Age r Value MN Protein NBUD carbonyl HDL LDL CRP WCC Lymp Neut CD4 CD8 CD4:8 Comet (fresh) Plasma Se NDI 20.29* NS 0.33** NS NS 0.29* NS NS NS NS NS NS NS NS MN — 0.34** NS NS NS 20.31* NS NS NS NS 0.31* NS NS NS NPB 0.43** NS NS NS NS NS NS NS NS NS NS NS NS NBUD — NS NS 0.27* NS NS NS NS NS NS NS NS NS F2 iso-prostanes NS 0.31* NS NS NS NS NS NS NS NS NS NS Protein carbonyl — NS NS NS NS NS NS NS NS NS 20.31* NS Chol NS 0.92** NS NS NS NS NS NS NS NS NS Trig 20.48** NS NS 0.27* NS NS 0.28* NS NS NS NS HDL — NS 20.34** 20.36** 20.26* 20.36** NS NS NS NS NS CRP — 0.34** 0.28* 0.34** NS NS NS NS 20.35** WCC — 0.70** 0.91** 0.34** NS 0.35** NS NS Lymp — 0.44** NS NS NS NS NS Neut — 0.37** NS 0.41** NS NS CD3 0.44** 0.47** NS NS NS CD4 — 20.51** 0.68** 0.38** NS CD8 — 20.79** 20.27* NS NS, nonsignificant, r values 2-tailed; NDI, nuclear division index; MN, binucleated cells with micronuclei; NPB, binucleated cells with nucleoplasmic bridge; NBUD, binucleated cells with nuclear bud; Chol, total plasma cholesterol; Trig, plasma triglyceride; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CRP, C-reactive protein; WCC, white cell count; Lymp, lymphocytes; Neut, neutrophils; CD3, CD3-positive lymphocytes; CD4, CD4-positive lymphocytes; CD8, CD8-positive lymphocytes; CD4:8, ratio of CD4-positive: CD8-positive lymphocytes; NS, nonsignificant. *P < 0.05; **P < 0.01. pleted the intervention in the CONTROL, BIOFORT, and PROFORT groups, respectively. Reasons for drop-out before commencement of the trial were increased work commitments and overseas travel. Reasons for drop-out during the trial included increased work commitments (N 5 3), overseas travel (N 5 1), voluntary withdrawal (N 5 2), increased family responsibilities (N 5 2), noncompliance (N 5 2), and one adverse event (N 5 1). Age (55.0 6 7.4 years, 56.1 6 Environmental and Molecular Mutagenesis. DOI 10.1002/em 6 Wu et al. TABLE IV. Frequencies of Binucleated Lymphocytes With Micronuclei (MN BNCs), Binucleated Lymphocytes With Nuclear Plasmic Bridges (NPB BNCs), Binucleated Lymphoctyes With Nuclear Buds (NBUD BNCs), and Nuclear Division Index (NDI) Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 7.9 6 4.0 9.8 6 5.8 8.0 6 6.2 CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 2.5 6 1.3 2.6 6 2.1 1.9 6 1.6 CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1.1 6 1.1 1.8 6 1.6 1.6 6 1.3 CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1.81 6 0.18 1.84 6 0.17 1.91 6 0.19 8 wk 16 wk MN BNCs per 1,000 cells 6.9 6 4.1 7.1 6 3.4 9.1 6 3.8 7.6 6 4.5 8.5 6 4.4 7.3 6 3.8 0.83 for treatment NPB BNCs per 1,000 cells 2.3 6 1.4 1.0 6 1.4 2.6 6 1.6 1.4 6 1.7 2.4 6 2.5 1.3 6 1.4 0.87 for treatment NBUD BNCs per 1,000 cells 1.3 6 2.0 0.8 6 1.1 1.7 6 1.8 0.8 6 1.1 1.3 6 1.6 0.7 6 1.0 0.60 for treatment NDI 1.82 6 0.14 1.80 6 0.19 1.88 6 0.19 1.91 6 0.19 1.83 6 0.16 1.94 6 0.13 0.59 for treatment 24 wk ANOVA Pa 6.8 6 4.3 8.2 6 5.3 8.1 6 3.7 0.62 for time 0.73 for treatment 3 time 0.7 6 0.9 0.8 6 1.2 1.3 6 1.1 0.07 for time 0.23 for treatment 3 time 0.6 6 1.1 0.8 6 1.1 0.6 6 1.0 0.51 for time 0.69 for treatment 3 time 1.89 6 0.13 1.93 6 0.23 1.97 6 0.17 0.59 for time 0.85 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P value in bold is statistically significant. TABLE V. DNA Damage (arbituary unit, AU) in Fresh Lymphocytes as Measured by the Comet Assay Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 35.4 6 12.8 30.7 6 15.1 27.4 6 11.0 CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 14.4 6 10.4 18.4 6 16.9 9.1 6 5.6 8 wk 16 wk Comet DNA damage, Category 0–3 (AU) 46.9 6 14.6 45.0 6 12.5 43.7 6 14.5 51.7 6 12.8 42.6 6 11.7 51.7 6 15.5 0.15 for treatment Comet DNA damage, category 4 (AU) 30.4 6 13.2 52.5 6 29.3 32.0 6 13.6 56.4 6 18.3 26.4 6 13.5 45.7 6 18.8 0.72 for treatment 24 wk ANOVA Pa 32.4 6 9.4 25.8 6 9.3 28.5 6 9.9 0.01 for time 0.39 for treatment 3 time 18.6 6 10.4 13.3 6 9.1 17.0 6 12.0 <0.0001 for time 0.13 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P values in bold are statistically significant. 6.6 years, and 55.8 6 7.1 years) or body mass index (BMI, 27.0 6 3.5 m2/kg, 26.0 6 3.3 m2/kg, 27.2 6 3.9 m2/kg) of men (data shown are mean 6 SD), the proportion of smokers (18.2, 10.5, and 28.6%) and alcohol consumers (86.4, 94.7, and 66.7%) was not significantly different between CONTROL, BIOFORT, and PROFORT groups, respectively. Baseline plasma Se concentrations in the three groups were not significantly different and the mean value (6SD) for CONTROL, BIOFORT, and PROFORT groups was 121.0 (68.8) lg/L, 122.3 (617.1) lg/L, and 122.3 (612.8) lg/L, respectively. The estimated Se intake was not different between CONTROL, BIOFORT, and PROFORT groups with the mean (6SD) intake of 155.5 (639.6) lg/day, 143.3 (681.8) lg/day, and 174.2 (656.0) lg/day, respectively. Compliance rate was high (>97%) in all groups. There was no change in the plasma Se concentration in the CONTROL group. However, in both the BIOFORT and PROFORT groups there was a significant time and dose-related increase in plasma Se (time effect P 5 0.007, treatment effect P 5 0.006), but the increment was much greater for the BIOFORT group in which a maximum increase of 70 lg/L Se was achieved by the end of the trial compared to an increase of 16 lg/L in the PROFORT group (Fig. 2). Despite the changes in plasma Se concentration, there was no significant effect on platelet Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers 7 TABLE VI. Plasma F2 Isoprostanes (pmol/L), Plasma Protein Carbonyl (nmol/mg protein) and Plasma C-Reactive Protein (mg/L) Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1865 (459) 1774 (495) 1805 (332) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 0.046 (0.022) 0.043 (0.047) 0.057 (0.058) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 2.24 (1.93) 1.76 (1.07) 1.66 (1.72) 8 wk 16 wk Plasma F2 isoprostanes (pmol/L) 1837 (428) 1930 (565) 1851 (517) 1852 (471) 1885 (438) 1834 (403) 0.621 for treatment Plasma protein carbonyl (nmol/mg protein) 0.043 (0.030) 0.046 (0.037) 0.058 (0.058) 0.050 (0.031) 0.049 (0.026) 0.069 (0.055) 0.174 for treatment Plasma C-reactive protein (mg/L) 1.85 (1.25) 1.86 (1.19) 1.57 (1.31) 2.12 (1.96) 1.71 (1.601) 1.70 (1.25) 0.391 for treatment 24 wk 1840 (430) 1880 (550) 1725 (388) ANOVA Pa 0.421 for time 0.202 for treatment 3 time 0.024 (0.015) 0.036 (0.032) 0.030 (0.018) 0.004 for time 0.206 for treatment 3 time 2.59 (1.97) 1.61 (1.35) 2.00 (2.52) 0.091 for time 0.187 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P value in bold is statistically significant. TABLE VII. Plasma Low-Density Lipoprotein (LDL) Cholesterol, Plasma High-Density Lipoprotein (HDL) Cholesterol, Plasma Triglycerides, and Plasma Total Cholesterol Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 3.62 (0.95) 3.32 (0.59) 3.14 (0.96) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1.29 (0.32) 1.42 (0.37) 1.21 (0.39) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1.58 (0.70) 1.38 (0.55) 2.19 (1.58) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 5.62 (0.98) 5.35 (0.67) 5.33 (1.05) 8 wk 16 wk Plasma LDL cholesterol (mmol/L) 3.56 (0.99) 3.57 (1.02) 3.39 (0.64) 3.52 (0.66) 3.15 (0.78) 3.20 (0.76) 0.020 for treatment Plasma HDL cholesterol (mmol/L) 1.28 (0.28) 1.26 (0.28) 1.41 (0.37) 1.43 (0.28) 1.17 (0.36) 1.22 (0.38) 0.779 for treatment Plasma triglycerides (mmol/L) 1.75 (0.77) 1.70 (0.85) 1.49 (0.43) 1.31 (0.36) 2.44 (1.97) 2.11 (1.41) 0.400 for treatment Plasma total cholesterol (mmol/L) 5.63 (1.12) 5.60 (1.11) 5.47 (0.78) 5.45 (0.69) 5.42 (0.95) 5.37 (0.69) 0.031 for treatment 24 wk ANOVA Pa 3.43 (0.92) 3.42 (0.51) 3.20 (0.89) 0.712 for time 0.958 for treatment 3 time 1.30 (0.27) 1.46 (0.35) 1.20 (0.36) 0.146 for time 0.310 for treatment 3 time 1.54 (0.68) 1.54 (0.52) 1.82 (0.90) 0.864 for time 0.342 for treatment 3 time 5.42 (0.94) 5.57 (0.62) 5.23 (0.91) 0.119 for time 0.674 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P values in bold are statistically significant. GPx activity (Fig. 3). A stratified analysis examining whether GPx values were increased following consumption of BIOFORT wheat found no effect of baseline GPx values on outcome (data not shown). The CBMN Cyt assay results (Table IV) showed no significant effect of time or treatment on frequency of MN BNCs (Fig. 4A), and this lack of effect of treatment was consistent in both those with an initially high and an initially low MN-BNC frequency at baseline (data not shown). Frequency of NPB BNCs and NBUD BNCs tended to decrease and NDI tended to increase (nonsignificantly) with time but there was no difference between treatments (Figs. 4B–4D). The Comet assay results (Table V) for both Category 0–3 and Category 4 showed a significant effect of time (P < 0.05) for fresh lymphocytes (Fig. 5) but not of treatment indicating a null effect of selenium supplementation on DNA damage as measured by the Comet assay. The Environmental and Molecular Mutagenesis. DOI 10.1002/em 8 Wu et al. TABLE VIII. White Cell Count, Lymphocyte Count, and Neutrophil Count Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 5.69 (1.67) 5.41 (1.19) 5.96 (1.47) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 1.751 (0.4763) 1.547 (0.2741) 2.009 (0.5862) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 3.43 (1.30) 3.38 (1.18) 3.47 (0.94) 8 wk 16 wk White cell count (3109/L) 6.08 (1.25) 5.83 (1.74) 5.39 (1.25) 5.48 (1.20) 6.42 (1.72) 6.29 (1.68) 0.876 for treatment Lymphocyte count (3109/L) 1.857 (0.5238) 1.841 (0.5741) 1.502 (1.906) 1.644 (0.4113) 2.069 (0.6145) 2.039 (0.6731) 0.106 for treatment Neutrophil count (3109/L) 3.66 (0.80) 3.44 (1.26) 3.31 (1.03) 3.35 (1.10) 3.86 (1.18) 3.76 (1.10) 0.080 for treatment 24 wk 5.90 (1.26) 5.66 (1.75) 6.39 (2.14) ANOVA Pa 0.234 for time 0.160 for treatment 3 time 1.759 (0.5427) 1.731 (0.5) 2.042 (0.6304) 0.228 for time 0.116 for treatment 3 time 3.60 (0.92) 3.60 (1.16) 3.86 (1.53) 0.043 for time 0.014 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P values in bold are statistically significant. TABLE IX. Percentage of CD3-Positive Lymphocytes, Percentage of CD4-Positive Lymphocytes, and Percentage of CD8-Positive Lymphocytes Study group Baseline CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 72.4 (5.8) 69.0 (8.4) 74.2 (6.1) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 47.5 (7.8) 46.5 (8.5) 49.4 (8.0) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 22.7 (8.9) 20.6 (8.5) 23.0 (8.2) CONTROL N 5 22 BIOFORT N 5 19 PROFORT N 5 21 ANOVA Pa 2.50 (1.31) 2.86 (1.72) 2.52 (1.18) 8 wk 16 wk Percentage of CD3-positive lymphocytes 72.2 (6.4) 71.2 (6.8) 70.1 (9.5) 68.9 (9.2) 75.0 (5.1) 73.6 (6.2) 0.456 for treatment Percentage of CD4-positive lymphocytes 48.1 (8.0) 46.4 (7.4) 46.6 (10.2) 45.7 (9.6) 50.2 (7.6) 49.7 (7.6) 0.422 for treatment Percentage of CD8-positive lymphocytes 22.0 (8.7) 22.6 (8.8) 21.2 (8.5) 20.2 (8.2) 23.3 (8.4) 22.4 (8.1) 0.435 for treatment Ratio of CD4-positive: CD8-positive lymphocytes 2.64 (1.38) 2.49 (1.34) 2.74 (1.66) 2.79 (1.57) 2.52 (1.19) 2.72 (1.58) 0.146 for treatment 24 wk ANOVA Pa 70.6 (7.1) 69.2 (8.8) 74.2 (5.5) 0.199 for time 0.692 for treatment 3 time 46.9 (6.8) 45.4 (9.1) 49.0 (7.4) 0.180 for time 0.487 for treatment 3 time 21.6 (6.8) 20.8 (8.0) 23.4 (7.9) 0.001 for time 0.032 for treatment 3 time 2.53 (1.38) 2.67 (1.50) 2.64 (1.82) 0.707 for time 0.387 for treatment 3 time a The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results shown are mean (SD). P values in bold are statistically significant. time effect seemed to be attributable to an increase in DNA damage at week 16 across all groups. Similar trends were observed for the Comet assay results with cryopreserved lymphocytes (data not shown). The ANOVA P value results for the secondary outcome measures are summarized in Table I and results detailed in Tables VI–IX. There was no significant effect of time or treatment on F2-isoprostanes. In contrast, plasma protein carbonyls exhibited a significant effect of time (P 5 0.004) but not of treatment with a marked reduction at week 24. CRP, leukocyte counts, and lymphocyte subsets were unaffected by the intervention except for minor effects of time on neutrophil and CD8 lymphocyte count. There was no significant difference between treatment groups with respect to plasma B12 and plasma folate during the intervention, but there was a marginal trend toward an increase in plasma folate concentration with time which achieved statistical signifi- Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers 9 tein (HDL, r 5 20.34, P 5 0.01) and HDL and triglyceride (r 5 20.48, P 5 0.01). DISCUSSION Fig. 2. Plasma Se concentration during the intervention trial. N 5 22, 19, 21 for CONTROL, BIOFORT, and PROFORT groups, respectively. The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results were mean 6 SEM (>). Fig. 3. Platelet glutathione peroxidase (GPx) activity during the intervention period. N 5 22, 19, 21 for CONTROL, BIOFORT, and PROFORT groups, respectively. The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results were mean 6 SEM (>). NS, nonsignificant. cance. BMI did not change significantly throughout the intervention for all groups (data not shown). A cross correlation analysis of the relationship between plasma Se (at the start and the end of the trial), after controlling for age, and all biomarkers in this study was also performed and found plasma Se was associated with CRP (r 5 20.35, P 5 0.01) and CD3 and CD4 lymphocyte count (r 5 20.31, 0.27, both P 5 0.05) but not other biomarkers (Tables II and III). Other relevant correlations were those between NDI and MN (r 5 20.29, P 5 0.05), and NBUD with MN and NPB (r 5 0.42, 0.43, both P 5 0.01) as well as CRP and high-density lipopro- The results of this study show that organic Se in biofortified wheat is bioavailable in a time and dose-related manner but has no impact on the health status biomarkers in the cohort studied. Although we selected participants with the lowest plasma Se concentrations in this cohort, the mean plasma Se concentration at baseline was 122 lg/L which is considered, in other studies in other countries, to be an adequate level to optimize function of biomarkers of selenium status such as GPx and in terms of minimizing prostate cancer risk attributable to selenium deficiency [Combs, 2005; Rayman, 2005; Brinkman et al., 2006; Gromadzinska et al., 2008]. Despite an increase of plasma Se concentration in the BIOFORT group to a maximum of 193 lg/L, there was no improvement in platelet GPx activity. Our results are consistent with those of previous studies showing that GPx activity in platelets and hemolysate is optimized at plasma Se concentration greater than 100–120 lg/L [Misso et al., 1996; Karunasinghe et al., 2006]. Similarly, there was no reduction in lymphocyte DNA damage measured using the CBMN Cyt assay or the Comet assay that could be explained by supplementation with the Se-biofortified wheat. Our in vitro studies showed that the CBMN Cyt assay is sensitive to both the cytoprotective/genome protective and cytotoxic/genotoxic effects of the organic form of Se, Se-met, and for this reason, as well as its comprehensive assessment of chromosome damage events is an ideal tool to investigate potential beneficial and adverse effects of food supplements and define simultaneously both bioefficacy and safety limits [Wu et al., in press]. The similar trend with time, but not treatment, for a reduction in plasma protein carbonyls suggests a reduction in oxidative stress occurring during the intervention in all groups in the aqueous phase which however was not reflected in the lipid phase given that F2-isoprostanes were unchanged during the intervention. However, changes in plasma protein carbonyl were not correlated with Comet or CBMN Cyt assay results (data not shown) suggesting that any variation in oxidative stress was below the threshold required to cause genome damage. The observed variation with time in the Comet assay could be due to unknown seasonal changes in exposure to genotoxic environmental factors because DNA damage in the Comet assay has been shown to vary significantly during the year and associated with increased exposure to sunlight, environmental pollutants such as polycyclic aromatic hydrocarbons, changes in exercise level and diet [Verschaeve et al., 2007]. The trends for reduction in Environmental and Molecular Mutagenesis. DOI 10.1002/em 10 Wu et al. Fig. 4. DNA damage and cytotoxicity measured using the CBMN Cyt assay. (A) Frequency of MN BNCs; (B) frequency of NPB BNCs; (C) frequency of NBUD BNCs; (D) NDI. N 5 22, 19, 21 for CONTROL, BIOFORT, and PROFORT groups, respectively. The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results were mean 6 SEM (>). NS, nonsignificant. some of the chromosome aberration biomarkers (NPB BNCs, NBUD BNCs) and protein carbonyls with time may indicate a seasonal variation in these biomarkers given that the intervention was of 6 months duration starting in late February/early March (mid/late summer in Australia) and ending in late August/early September (mid/late winter) that could be related to environmental factors such as exposure to UV radiation and/or heat stress as well as change in diet which is suggested by the increase in plasma folate with time. The trend for reduced NPB BNCs and NBUD BNCs with time may be explained by the increased folate status because these biomarkers have been shown to be sensitive to folate concentration [Crott et al., 2001; Kimura et al., 2004]. However, this hypothesis is negated by the fact that MN frequency, which is also folate-sensitive, was unaffected by season which agrees with observations from our laboratory in a previous study [Fenech, 1998]. Se supplementation clearly had no marked beneficial impact on leucocyte count or ratios, suggesting that immune function was not substantially affected. This is also supported by the lack of a significant impact on NDI which is a measure of mitogenic response of lymphocytes and thus considered a surrogate marker of immune response. We have considered the possibility of measuring oxidized DNA bases however we came to the conclusion that the combination of the Comet and the CBMN Cyt assay would be sufficient to indirectly also detect DNA base damage by oxidative stress. This is because reactive oxygen species (ROS)-induced DNA base damage leads to the formation of abasic sites during base-excision repair which are detectable in the alkaline Comet assay and MN and NPB in the CBMN Cyt assay are efficiently induced by ROS such as H2O2, superoxide, and activated neutrophils [Umegaki and Fenech, 2000; Dotan et al., 2004; Muth et al., 2004; Devaraj et al., 2008]. Furthermore, we felt that it was im- Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers 11 markers of this disease such as lipid profile and lipid oxidation. There was a weak negative correlation with CRP at baseline and at the end of the trial but no impact of Se supplementation on this biomarker of inflammation and cardiovascular disease. Our results are in agreement with those of Ravn-Haren et al. [2008] who also showed no benefit of Se supplementation on conventional biomarkers of cardiovascular disease. In another study in patients with coronary artery disease there was no impact of Se supplementation (as selenite) on endothelial function, biomarkers of inflammation, or oxidative stress even though GPx activity was increased [Schnabel et al., 2008]. CONCLUSIONS Fig. 5. DNA damage in fresh lymphocytes measured using the Comet assay. (A) Results for cells showing category 0–3 DNA damage levels. (B) Results for cells showing category 4 DNA damage. N 5 22, 19, 21 for CONTROL, BIOFORT, and PROFORT groups, respectively. The statistical analysis was performed using delta value of result of each follow-up time point relative to baseline result and baseline value was included as covariate using mixed between-within subjects ANOVA. Results were mean 6 SEM (>). NS, nonsignificant. portant to use any additional effort and resources to measure other biomarker of oxidative stress such as protein carbonyls and F2-isoprostanes because it is evident that measurement with DNA end-points alone is not sufficient to obtain the full spectrum of ROS-induced damage to the organism [Dotan et al., 2004; Muth et al., 2004; Devaraj et al., 2008]. We did not conduct the challenge assay with ROS because our in vitro studies with human peripheral blood lymphocytes with Se-met, the predominant organic form of selenium, covering both physiological and supraphysiological concentrations, showed no impact on ionizing radiation-induced DNA damage and chromosome aberrations measured using the Comet and the CBMN Cyt assay, respectively [Wu et al., in press]. Despite the reported associations between increased Se intake with decreased cardiovascular disease risk, our study showed no beneficial impact on conventional bio- The results from this study show that (a) supplementation with dietary Se in older men with already replete plasma Se concentrations does not confer any additional health benefits and (b) increasing plasma Se concentration up to 193 lg/L using BIOFORT wheat does not appear to have any obvious toxic effects, within the limits of the study time-frame, the number of individuals in the trial, and the range of biomarkers measured. Therefore, dietary intake of 105–315 lg of organic selenium from BIOFORT wheat biscuits in addition to 158 lg from other habitual dietary sources (calculated using a validated food frequency questionnaire) in South Australia is unlikely to cause beneficial or adverse health outcomes in the short term based on the biomarkers of cancer and cardiovascular disease risk we used. These observations are in accord with results of the SELECT trial and the EPIC study [Allen et al., 2008; Lippman et al., 2009], which showed no evidence of a link between selenium status and prostate cancer risk in well-nourished men. Given that there is some evidence that the benefits and/or adverse effects of Se supplementation may depend on genotype variations due to common polymorphisms in key genes such as GPx, selenoprotein-P, and thioredoxin reductase-1 [Hu and Diamond, 2003; Hu et al., 2005; Cai et al., 2006; Foster et al., 2006], it will be necessary to find out in future studies whether there are any specific genotypes or higher susceptibility groups, including prostate cancer patients, who are more likely to benefit, or be at risk of toxic effects, from greater intake of wheat enriched in Se by agronomic biofortification. ACKNOWLEDGMENTS The authors are grateful to Dr. Kath Cooper, and Martin and Kirsty Flower for assistance with production of biofortified grain; Andrew Van der Sluys, Mark Laucke, and David Hogan from Laucke Flour Mills for biscuit for- Environmental and Molecular Mutagenesis. DOI 10.1002/em 12 Wu et al. mulation and production; Peter Babidge (South Australian Research & Development Institute) for fluorimetric analyses of selenium; Teresa Fowles, Lyndon Palmer, and staff at Waite Analytical Services for mass spectrometry and atomic emission spectrometry; Jason Kirby (CSIRO Land and Water, Adelaide) for performing the Se speciation measurements; Professor Peter McLennan (University of Wollongong) and Dr. Alice Owen (Monash University) for protein carbonyl assays; Jonathan Hodgson and Kevin Croft (University of Western Australia) for F2-isoprostane assays; and Candita Sullivan for C-reactive protein assays. The authors are grateful to Kylie Lange for her role as biostatistician in providing advice on appropriate analyses of data; the participants and the staff at the CSIRO’s Clinical Trials Unit for recruitment of volunteers and management of the trial. REFERENCES Allen NE, Appleby PN, Roddam AW, Tjønneland A, Johnsen NF, Overvad K, Boeing H, Weikert S, Kaaks R, Linseisen J, Trichopoulou A, Misirli G, Trichopoulos D, Sacerdote C, Grioni S, Palli D, Tumino R, Bueno-de-Mesquita HB, Kiemeney LA, Barricarte A, Larrañaga N, Sánchez MJ, Agudo A, Tormo MJ, Rodriguez L, Stattin P, Hallmans G, Bingham S, Khaw KT, Slimani N, Rinaldi S, Boffetta P, Riboli E, Key TJ. 2008. Plasma selenium concentration and prostate cancer risk: Results from the European Prospective Investigation into Cancer and Nutrition (EPIC). Am J Clin Nutr 88:1567–1575. Baghurst KI, Record SJ. 1984. A computerised dietary analysis system for use with diet diaries or food frequency questionnaires. Community Health Stud 8:11–18. Baghurst PA, Carman JA, Syrette JA, Baghurst KI, Crocker JM. 1992. Diet, prolactin, and breast cancer. Am J Clin Nutr 56:943–949. Bjelakovic G, Nikolova D, Simonetti RG, Gluud C. 2004. Antioxidant supplements for prevention of gastrointestinal cancers: A systematic review and meta-analysis. Lancet 364:1219–1228. Bonassi S, Znaor A, Ceppi M, Lando C, Chang WP, Holland N, KirschVolders M, Zeiger E, Ban S, Barale R, Bigatti MP, Bolognesi C, Cebulska-Wasilewska A, Fabianova E, Fucic A, Hagmar L, Joksic G, Martelli A, Migliore L, Mirkova E, Scarfi MR, Zijno A, Norppa H, Fenech M. 2007. An increased micronucleus frequency in peripheral blood lymphocytes predicts the risk of cancer in humans. Carcinogenesis 28:625–631. Brinkman M, Reulen RC, Kellen E, Buntinx F, Zeegers MP. 2006. Are men with low selenium levels at increased risk of prostate cancer? Eur J Cancer 42:2463–2471. Buss H, Chan TP, Sluis KB, Domigan NM, Winterbourn CC. 1997. Protein carbonyl measurement by a sensitive ELISA method. Free Radic Biol Med 23:361–366. Cai L, You NC, Lu H, Mu LN, Lu QY, Yu SZ, Le AD, Marshall J, Heber D, Zhang ZF. 2006. Dietary selenium intake, aldehyde dehydrogenase-2 and X-ray repair cross-complementing 1 genetic polymorphisms, and the risk of esophageal squamous cell carcinoma. Cancer 106:2345–2354. Collins A. 2002. The Comet assay—Principles, applications, and limitations. In: Didenko VV, editor. Methods in Molecular Biology, In Situ Detection of DNA Damage, Methods and Protocols. Totowa, NJ:Humana Press. pp 163–179. Collins AR. 2005. Assays for oxidative stress and antioxidant status: Applications to research into the biological effectiveness of polyphenols. Am J Clin Nutr 81(1Suppl):261S–267S. Combs G Jr. 2005. Current evidence and research needs to support a health claim for selenium and cancer prevention. J Nutr 135:343–347. Crott JW, Mashiyama ST, Ames BN, Fenech M. 2001. The effect of folic acid deficiency and MTHFR C677T polymorphism on chromosome damage in human lymphocytes in vitro. Cancer Epidemiol Biomarkers Prev 10:1089–1096. Devaraj S, Mathur S, Basu A, Aung HH, Vasu VT, Meyers S, Jialal I. 2008. A dose-response study on the effects of purified lycopene supplementation on biomarkers of oxidative stress. J Am Coll Nutr 27:267–273. Dotan Y, Lichtenberg D, Pinchuk I. 2004. Lipid peroxidation cannot be used as a universal criterion of oxidative stress. Prog Lipid Res 43:200–227. El-Zein RA, Fenech M, Lopez MS, Spitz MR, Etzel CJ. 2008. Cytokinesis-blocked micronucleus cytome assay biomarkers identify lung cancer cases amongst smokers. Cancer Epidemiol Biomarkers Prev 17:1111–1119. Fenech M. 1998. Important variables that influence base-line micronucleus frequency in cytokinesis-blocked lymphocytes—A biomarker for DNA damage in human populations. Mutat Res 404:155–165. Fenech M. 2007. Cytokinesis-block micronucleus cytome assay. Nat Protoc 2:1084–1104. Foster CB, Aswath K, Chanock SJ, McKay HF, Peters U. 2006. Polymorphism analysis of six selenoprotein genes: Support for a selective sweep at the glutathione peroxidase 1 locus (3p21) in Asian populations. BMC Genet 7:56. Gromadzinska J, Reszka E, Bruzelius K, Wasowicz W, Akesson B. 2008. Selenium and cancer: Biomarkers of selenium status and molecular action of selenium supplements. Eur J Nutr 47(Suppl 2):29–50. Gromer S, Eubel JK, Lee BL, Jacob J. 2005. Human selenoproteins at a glance. Cell Mol Life Sci 62:2414–2437. Gronberg H. 2003. Prostate cancer epidemiology. Lancet 361:859–864. Hatfield DL, Carlson BA, Xu XM, Mix H, Gladyshev VN. 2006. Selenocysteine incorporation machinery and the role of selenoproteins in development and health. Prog Nucleic Acid Res Mol Biol 81:97–142. Hu Y, Benya RV, Carroll RE, Diamond AM. 2005. Allelic loss of the gene for the GPX1 selenium-containing protein is a common event in cancer. J Nutr 135(12Suppl):3021S–3024S. Hu YJ, Diamond AM. 2003. Role of glutathione peroxidase 1 in breast cancer: Loss of heterozygosity and allelic differences in the response to selenium. Cancer Res 63:3347–3351. Karunasinghe N, Ryan J, Tuckey J, Masters J, Jamieson M, Clarke L, Marshall J, Ferguson L. 2004. DNA stability and serum selenium levels in a high-risk group for prostate cancer. Cancer Epidemiol Biomarkers Prev 13:391–397. Karunasinghe N, Ferguson LR, Tuckey J, Masters J. 2006. Hemolysate thioredoxin reductase and glutathione peroxidase activities correlate with serum selenium in a group of New Zealand men at high prostate cancer risk. J Nutr 136:2232–2235. Kimura M, Umegaki K, Higuchi M, Thomas P, Fenech M. 2004. Methylenetetrahydrofolate reductase C677T polymorphism, folic acid and riboflavin are important determinants of genome stability in cultured human lymphocytes. J Nutr 134:48–56. Kirby J, Lyons G, Markkainen M, McLaughlin M. HPLC-(ID)-ICPMS for the determination of selenium species in biofortified grains and biscuits. In: INTERACT 2006, September 24–28, Perth, Western Australia. Koh T-S, Benson T. 1983. Critical re-appraisal of fluorometric method for determination of selenium in biological materials. Anal Chem 66:918–926. Lippman SM, Klein EA, Goodman PJ, Lucia MS, Thompson IM, Ford LG, Parnes HL, Minasian LM, Gaziano JM, Hartline JA, Parsons JK, Bearden JD 3rd, Crawford ED, Goodman GE, Claudio J, Winquist E, Cook ED, Karp DD, Walther P, Lieber MM, Kristal Environmental and Molecular Mutagenesis. DOI 10.1002/em Se-Biofortified Wheat and Health Status Biomarkers AR, Darke AK, Arnold KB, Ganz PA, Santella RM, Albanes D, Taylor PR, Probstfield JL, Jagpal TJ, Crowley JJ, Meyskens FL Jr, Baker LH, Coltman CA Jr. 2009. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: The Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA 301:39–51. Lyons G, Stangoulis J, Graham R. 2003. High-selenium wheat: Biofortification for better health. Nutr Res Rev 16:45–60. Lyons G, Judson G, Stangoulis J, Palmer L, Jones J, Graham R. 2004. Trends in selenium status of South Australians. Med J Aust 180:383–386. Lyons G, Judson G, Ortiz-Monasterio I, Genc Y, Stangoulis J, Graham R. 2005a. Selenium in Australia: Selenium status and biofortification of wheat for better health. J Trace Elem Med Biol 19:75–82. Lyons G, Ortiz-Monasterio I, Stangoulis J, Graham G. 2005b. Selenium concentration in wheat grain: Is there sufficient genotypic variation to use in breeding? Plant Soil 269:369–380. Misso N, Powers K, Gillon R, Stewart G, Thompson P. 1996. Reduced platelet gutathione peroxidase activity and serum selenium concentration in atopic asthmatic patients. Clin Exp Allergy 26:838– 847. Mori TA, Croft KD, Puddey IB, Beilin LJ. 1999. An improved method for the measurement of urinary and plasma F2-isoprostanes using gas chromatography-mass spectrometry. Anal Biochem 268:117–125. Muth CM, Glenz Y, Klaus M, Radermacher P, Speit G, Leverve X. 2004. Influence of an orally effective SOD on hyperbaric oxygenrelated cell damage. Free Radic Res 38:927–932. Pallant J. 2005. SPSS survival manual: A Step by Step Guide to Data Analysis Using SPSS for Windows (Version 12). Crows Nest, NSW:Allen & Unwin. Piperakis SM, Petrakou E, Tsilimigaki S, Sagnou M, Monogiudis E, Haniotakis G, Karkaseli H, Sarikaki E. 2003. Biomonitoring with the comet assay of Greek greenhouse workers exposed to pesticides. Environ Mol Mutagen 41:104–110. Ravn-Haren G, Bugel S, Krath BN, Hoac T, Stagsted J, Jorgensen K, Bresson JR, Larsen EH, Dragsted LO. 2008. A short-term intervention trial with selenate, selenium-enriched yeast and seleniumenriched milk: Effects on oxidative defence regulation. Br J Nutr 99:883–892. Rayman M. 2000. The importance of selenium to human health. Lancet 356:233–241. 13 Rayman MP. 2005. Selenium in cancer prevention: A review of the evidence and mechanism of action. Proc Nutr Soc 64:527–542. Reid ME, Stratton MS, Lillico AJ, Fakih M, Natarajan R, Clark LC, Marshall JR. 2004. A report of high-dose selenium supplementation: Response and toxicities. J Trace Elem Med Biol 18:69–74. Schnabel R, Lubos E, Messow CM, Sinning CR, Zeller T, Wild PS, Peetz D, Handy DE, Munzel T, Loscalzo J, Lackner KJ, Blankenberg S. 2008. Selenium supplementation improves antioxidant capacity in vitro and in vivo in patients with coronary artery disease The SElenium Therapy in Coronary Artery disease Patients (SETCAP) Study. Am Heart J 156:1201.e1–1201.e11. Umegaki K, Fenech M. 2000. Cytokinesis-block micronucleus assay in WIL2-NS cells: A sensitive system to detect chromosomal damage induced by reactive oxygen species and activated human neutrophils. Mutagenesis 15:261–269. Verschaeve L, Koppen G, Gorp UV, Schoeters G, Jacobs G, Zwijzen C. 2007. Seasonal variations in spontaneous levels of DNA damage; implication in the risk assessment of environmental chemicals. J Appl Toxicol 27:612–620. Waters D, Shen S, Cooley D, Bostwick D, Qian J, Combs G Jr, Glickman L, Oteham C, Schlittler D, Morris J. 2003. Effects of dietary selenium supplementation on DNA damage and apoptosis in canine prostate. J Natl Cancer Inst 95:237–241. Waters DJ, Shen S, Glickman LT, Cooley DM, Bostwick DG, Qian J, Combs GF Jr, Morris JS. 2005. Prostate cancer risk and DNA damage: Translational significance of selenium supplementation in a canine model. Carcinogenesis 26:1256–1262. Watkinson J. 1966. Fluorometric determination of selenium in biological material with 2,3-diaminonaphthalene. Anal Chem 38:92–97. Wu J, Lyons GH, Graham RD, Fenech MF. The effect of selenium, as selenomethionine, on genome stability and cytotoxicity in human lymphocytes measured using the cytokinesis-block micronucleus cytome assay. Mutagenesis. 2009 Jan 20. [Epub ahead of print]. Yang G, Zhou R. 1994. Further observations on the human maximum safe dietary selenium intake in a seleniferous area of China. J Trace Elem Electrolytes Health Dis 8:159–165. Accepted by— H. Norppa