Eric O. Uthus

Determinants of selenium status in healthy adults

BackgroundSelenium (Se) status in non-deficient subjects is typically assessed by the Se contents of plasma/serum. That pool comprises two functional, specific selenoprotein components and at least one non-functional, non-specific components which respond differently to changes in Se intake. A more informative means of characterizing Se status in non-deficient individuals is needed.MethodsMultiple biomarkers of Se status (plasma Se, serum selenoprotein P [SEPP1], plasma glutathione peroxidase activity [GPX3], buccal cell Se, urinary Se) were evaluated in relation to selenoprotein genotypes (GPX1, GPX3, SEPP1, SEP15), dietary Se intake, and parameters of single-carbon metabolism in a cohort of healthy, non-Se-deficient men (n = 106) and women (n = 155).ConclusionsPlasma Se concentration was 142.0 ± 23.5 ng/ml, with GPX3 and serum-derived SEPP1 calculated to comprise 20% and 34%, respectively, of that total. The balance, comprised of non-specific components, accounted for virtually all of the interindividual variation in total plasma Se. Buccal cell Se was associated with age and plasma homocysteine (hCys), but not plasma Se. SEPP1 showed a quadratic relationship with body mass index, peaking at BMI 25-30. Urinary Se was greater in women than men, and was associated with metabolic body weight (kg0.75), plasma folate, vitamin B12 and hCys (negatively). One GPX1 genotype (679T/T) was associated with significantly lower plasma Se levels than other allelic variants. Selenium intake, estimated from food frequency questionnaires, did not predict Se status as indicated by any biomarker. These results show that genotype, methyl-group status and BMI contribute to variation in Se biomarkers in Se-adequate individuals.

Differential Effects of Dietary Selenium (Se) and Folate on Methyl Metabolism in Liver and Colon of Rats

A previous study compared the effects of folate on methyl metabolism in colon and liver of rats fed a selenium-deficient die (<3 μg Se/kg) to those of rats fed a diet containing supranutritional Se (2 mg selenite/kg). The purpose of this study was to investigate the effects of folate and adequate Se (0.2 mg/kg) on methyl metabolism in colon and liver. Weanling, Fischer-344 rats (n=8/diet) were fed diets containing 0 or 0.2 mg selenium (as selenite)/kg and 0 or 2 mg folic acid/kg in a 2×2 design. After 70 d, plasma homocysteine was increased (p<0.0001) by folate deficiency; this increase was markedly, attenuated (p<0.0001) in rats fed the selenium-deficient diet compared to those fed 0.2 mg Se/kg. The activity of hepatic glycine N-methyltransferase (GNMT), an enzyme involved in the regulation of tissue S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH), was increased by folate deficiency (p<0.006) and decreased by selenium deprivation, (p<0.0003). Colon and liver SAH were highest (p<0.006) in rats fed deficient folate and adequate selenium. Although folate deficiency decreased liver SAM (p<0.001), it had no effect on colon SAM. Global DNA methylation was decreased (p<0.04) by selenium deficiency in colon but not liver; folate had no effect. Selenium, deficiency did not affect DNA methyltransferase (Dnmt) activity in liver but tended to decrease (p<0.06) the activity of the enzyme in the colon. Dietary folate did not affect liver or colon Dnmt. These results in rats fed adequate selenium are similar to previous results found in rats fed supranutritional selenium. This suggests that selenium deficiency appears to be a more important modifier of methyl metabolism than either adequate or supplemental selenium.

Prenatal Low-Protein and Postnatal High-Fat Diets Induce Rapid Adipose Tissue Growth by Inducing Igf2 Expression in Sprague Dawley Rat Offspring

Maternal low-protein diets result in lower birth weight followed by accelerated catch-up growth that is accompanied by the development of obesity and glucose intolerance in later life. Whether postnatal high-fat (HF) diets further contribute to the development of obesity and insulin resistance in offspring by affecting adipose tissue metabolism and DNA methylation is currently unknown. Obese-prone Sprague-Dawley rats were fed 8% low protein (LP) or 20% normal protein diets for 3 wk prior to conception and throughout pregnancy and lactation to investigate whether prenatal LP and postnatal HF diets affect the rate of adipose tissue growth, insulin-like growth factor 2 (Igf2) expression, and DNA methylation in male offspring. At weaning, the offspring were fed 10% normal fat or 45% HF diets for 12 wk. The adipose tissue growth rate was increased (up to 26-fold) by the LP prenatal and HF postnatal diets. Adipose tissue Igf2 mRNAs and DNA methylation were increased by the LP prenatal and HF postnatal diets. The LP prenatal and HF postnatal diet increased the number of small adipocytes in adipose tissue and decreased insulin sensitivity. These findings suggest that prenatal LP and postnatal HF intake result in adipose tissue catch-up growth through alterations in the expression of the Igf2 gene and DNA methylation within adipocytes. These alterations in adiposity are accompanied by an increased risk of development of type 2 diabetes.

Dietary Selenomethionine Increases Exon-Specific DNA Methylation of the p53 Gene in Rat Liver and Colon Mucosa

The regulation of site-specific DNA methylation of tumor suppressor genes has been considered as a leading mechanism by which certain nutrients exert their anticancer property. This study was to investigate whether selenium (Se) affects the methylation of globe genomic DNA and the exon-specific p53 gene. Three groups of rats (n = 6-7/group) were fed the AIN-93G basal diet supplemented with 0 [Se deficient (D)], 0.15 [Se adequate (A)], or 4 mg [Se supranutritional (S)] (Se as l-selenomethionine)/kg diet for 104 d, respectively. Rats fed the A or S diet had greater plasma and liver glutathione peroxidase activity, liver thioredoxin reductase activity, and plasma homocysteine concentration than those fed the D diet. However, compared with the A diet, rats fed the S diet did not further increase these Se-dependent enzyme activities or homocysteine concentration. In contrast, Se concentrations in kidney, liver, gastrocnemius muscle, and plasma were increased in a Se-dose-dependent manner. Interestingly, rats fed the S diet had significantly less global liver genomic DNA methylation than those fed the D diet. However, the S diet significantly increased the methylation of the p53 gene (exons 5-8) but not the β-actin gene (exons 2-3) DNA in liver and colon mucosa compared with those fed the D diet. Taken together, long-term Se consumption not only affects selenoprotein enzyme activities, homocysteine, tissue Se concentrations, and global genomic DNA methylation but also increases exon-specific DNA methylation of the p53 gene in a Se-dose-dependent manner in rat liver and colon mucosa.

First trimester vitamin D status and placental epigenomics in preeclampsia among Northern Plains primiparas.

AIMS As maternal vitamin D status has been associated with preeclampsia, the purpose of this study was to determine variations in DNA methylation patterns and associated protein expression in placental genes regulating vitamin D metabolism. MAIN METHODS A convenience sample of 48 pregnant nulliparous women, including 11 later diagnosed with preeclampsia, were recruited in this prospective study. Using a case-control design in two groups of women, we administered a food frequency questionnaire to determine vitamin D dietary intake. Laboratory measures included serum vitamin D levels (25[OH]D), DNA methylation patterns and protein expression in placental genes regulating vitamin D metabolism (1α-hydroxylase, CYP27B1; vitamin D receptor, VDR; retinoid X receptor, RXR) from placental tissue collected at delivery among those diagnosed with preeclampsia and those who remained normotensive throughout pregnancy. KEY FINDINGS There were no significant differences in vitamin D dietary intake or mean serum 25[OH]D levels, although the proportion of women with deficient 25[OH]D levels was higher in the preeclampsia group (46%) than the normotensive group (20%). Placenta samples from women with preeclampsia also had increased DNA methylation of CYP27B1, VDR and RXR genes with lower protein expression levels limited to RXR. SIGNIFICANCE Hypermethylation of key placental genes involved in vitamin D metabolism suggests uncoupling of processes that may interfere with placentation and availability of vitamin D at the maternal-fetal interface.

Decreased beige adipocyte number and mitochondrial respiration coincide with increased histone methyl transferase (G9a) and reduced FGF21 gene expression in Sprague-Dawley rats fed prenatal low protein and postnatal high-fat diets.

We have shown that prenatal low-protein (LP) followed by postnatal high-fat (HF) diets result in a rapid increase in subcutaneous adipose tissue (subc-AT) mass in the offspring, contributing to development of obesity and insulin resistance. Studies have shown that a key transcription factor, PR domain containing 16 (PRDM16), and fibroblast growth factor 21 (FGF21) are involved in conversion of precursor cells into mitochondria (mt)-enriched beige adipocytes (BA). Our hypothesis is that a maternal LP and postnatal HF diets increase the risk of obesity and insulin resistance in offspring, in part, by reducing the conversion of precursor cell into BA in the subc-AT of offspring. Using obese-prone Sprague-Dawley rats fed 8% LP or 20% normal-protein (NP) diets for 3 weeks prior to conception and throughout pregnancy and lactation followed by 12 weeks of 10% normal-fat (NF) or 45% HF diet feeding, we investigated whether prenatal LP and postnatal HF diets affect BA number and oxidative respiratory function in subc-AT. Results showed that subc-AT and liver FGF21, PRDM16 and BA marker CD137 mRNA increase with postnatal HF diet in maternal NP group rats. In contrast, rats fed maternal LP and postnatal HF diets showed no increase in subc-AT mt copy number, oxygen consumption rate, FGF21, PRDM16 and CD137 mRNA, whereas protein expression of an inhibitor for FGF21 transcription (histone methyltransferase, G9a) increased. These findings suggest that LPHF diets cause offspring metabolic alterations by reduced BA and FGF21 mRNA and increased G9a protein expression in subc-AT.

Interaction Between Zinc and Iron in Rats Experimental Results and Mathematical Analysis of Blood Parameters

The importance of interactive effects, of minerals in general, on nutrient requirements is becoming increasingly recognized. The interaction between iron and zinc has not been as widely investigated. The metabolic interrelationships between dietary iron and zinc have been known for years, but some subtle relationships may have gone unrecognized. Because nutrient interactions are not necessarily linear in nature, it may be inadequate to apply linear statistical models to study the interaction between zinc and iron. In this study, we used traditional as well as a nonlinear approach in analyzing experimental results from groups of rats fed a wide range of dietary zinc and iron. Male weanling Sprague-Dawley rats were used in a 5 × 4 factorially arranged experiment. Dietary variables were iron (as ferric citrate) at 4, 12, 24, 48, or 96 µg Fe/g diet and zinc (as zinc carbonate) at 5, 10, 20, or 40 µg Zn/g diet. After 7 wk, hematological parameters were measured and plasma ceruloplasmin and cholesterol were determined. In addition to interactive effects as shown by analysis of variance, the application of log-linear analysis to the experimental data revealed a far broader range of interactions between dietary iron and zinc. As a result of our experiment and its quantitative analysis, we conclude that the interaction between iron and zinc is nutritionally important and that dietary iron affected the response of many blood parameters to dietary zinc. The complete dataset can be found at http://www.gfhnrc.ars.usda.gov/fezn.

Effects in rats of iron on lead deprivation

In two fully crossed, two-factor experiments, F1 generation male rats were fed a basal diet supplemented with lead (lead acetate) at 0 or 2 μg/g and iron (ferric sulfate) at 50 or 250 μg/g (Experiment 1). Supplements in Experiment 2 were lead at 0 or 1 μg/g and iron at 50, 250, or 1000 μg/g. After 28 or 50 d in Experiment 1, and 35 d in Experiment 2, a relationship between lead and iron was found. Body weight was lower in low-lead than lead-supplemented 28-d-old rats regardless of dietary iron, whereas hematocrit and hemoglobin were lower in low-lead than lead-supplemented rats fed 50 μg iron/g diet. A similar finding was obtained with hematocrit and hemoglobin in 35-d-old rats. Dietary lead did not affect rats fed 250 or 1000 μg iron/g diet. Also, feeding low dietary lead did not affect 50-d-old rats regardless of dietary iron. Liver and bone concentrations of lead were markedly affected by dietary lead and iron. The concentration of lead in liver and bone was lower in low-lead than lead-supplemented rats. Compared to rats fed 50 μg iron/g diet, rats fed 250 μg iron/g diet exhibited a decreased lead concentration in liver and bone. This decrease was accentuated by lead supplementation. The findings suggest that lead acted pharmacologically to affect iron metabolism in rats.

Effect of dietary nickel deprivation on vision, olfaction, and taste in rats

Early studies on dietary nickel deprivation found decreased reproduction rate in pigs and decreased insemination and conception rates in goats. Studies from our laboratory demonstrated that nickel deprivation impaired male reproductive function of rats. A physiological amount of nickel modulates the function of cyclic nucleotide-gated cation channels (CNG channels) in vitro. Thus, because CNG channels have important roles in spermatozoa function, it was speculated that the impairment of reproduction by nickel deprivation was through an effect on CNG channels. Because CNG channels are found in retinal photoreceptor, olfactory receptor, and taste receptor cells, we hypothesized that nickel deprivation would also alter light/dark preference, odor preference to female rat urine, and taste preference/aversion in rats. In the light/dark Y-maze, nickel deprivation significantly decreased time spent in the dark arm by rats. The number of sniffs to estrous female urine was significantly increased only in nickel-supplemented rats. The number of licks at the saccharin bottle was significantly decreased by dietary nickel deprivation. These findings suggest that nickel has a biological role in the special senses: vision, olfaction and taste.

Validation of DNA Methylation Patterns Potential Biomarker for Heritable Risk of Preeclampsia

The purpose of this study was to identify and validate specific epigenetic DNA methylation patterns regulating gene expression in women with preeclampsia (PE). PE is a pregnancy-specific condition that causes complications during pregnancy, and increases future risk of developing hypertension and PE in offspring. There are no known preventive or early screening methods for PE. Detection of PE prior to symptom onset may be possible utilizing epigenetic biomarkers. David Barker’s theory of Developmental Origins of Health and Disease postulates fetal programming, predisposing individuals to illnesses later in life, occurs in utero. Maternal peripheral blood cells (MPBCs) and placental (fetal-origin) tissue samples from women with and without PE (n = 6/group) were analyzed for differences in DNA methylation in individual CpG dinucleotides. Delta beta criteria of >0.2 or <−0.2 between samples were used to determine significant gain or loss of methylation. Gene selection for validation was based on function and common patterns of methylation between sample types. Validation of methylation patterns was completed using bisulfite sequencing and pyrosequencing. Methylation increased in 123 MPBC and 63 placental genes, and decreased in 67 MPBC and 69 placental genes. Functional genes for lipid metabolism, inflammation, and vascular function were identified. Validation of methylation was confirmed on 7 candidate genes common to MPBCs and placenta samples. Differentially methylated genes common to maternal and fetal tissue represent a putative mechanism for perpetuated risk for PE from mother to offspring. Screening using a select panel of PE-associated epigenetic methylation patterns in validated genes has the potential to be a sensitive and specific clinical screening tool to identify risk for developing PE and hypertension. Early detection would allow targeted prevention and treatment improving perinatal and cardiovascular health.