| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* Institute for Cell and Molecular Biosciences,
Human Nutrition Research Centre, and
School of Clinical Medical Sciences, The Medical School, Newcastle University, Newcastle-upon-Tyne, UK;
Department of Clinical Biochemistry, University of Edinburgh, Edinburgh, UK;
¶ Division of Gastroenterology, Hepatology, and Nutrition, Department of Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee, USA;
|| BioSS, Aberdeen, UK; and
# Rowett Research Institute, Bucksburn, Aberdeen, UK
1Correspondence: Institute for Cell and Molecular Biosciences and Human Nutrition Research Centre, The Medical School, Newcastle University, Newcastle-upon-Tyne, NE2 4HH UK. E-mail: j.e.hesketh{at}ncl.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: nutrient-gene interaction • 3'untranslated region • BMI
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. It exerts many biological functions as the amino acid selenocysteine (Sec), which is present in a family of proteins called selenoproteins (e.g., glutathione peroxidases, thioredoxin reductases, selenoprotein P (SePP), and iodothyronine deiodinases). Selenoproteins are involved in the control of the cell redox status, immune function, Se metabolism, and thyroid hormone metabolism (2
, 3)
. During selenoprotein synthesis, Sec is incorporated at a recoded UGA codon, which normally signals the termination of translation (4
, 5)
. This process involves the formation of complexes with a RNA stem-loop structure [selenocysteine-insertion sequence (SECIS)] located in the 3'untranslated region (3'UTR) of selenoprotein mRNA. Adequate Se intake is required for full expression of selenoproteins (6
, 7)
.
Several studies (1
, 8
, 9)
have suggested a chemopreventive effect of Se supplementation, whereas suboptimal Se status has been associated with increased susceptibility to diseases notably cancers and cardiovascular disorder. Recently, two prospective studies have associated low plasma Se status with increased rates of mortality: in a French study of both men and women [Etude du Viellissement Artériel (EVA)] low Se status was correlated with increased risk of death from cancer (10)
, and similar results were obtained in the USA in the Women’s Health and Aging Study in which women with higher Se and carotenoids levels were at lower risk of death (11)
. Despite such studies, the definition of a healthy Se status is still a matter of debate (8)
. At present, the UK recommended daily Se intake is 55 µg/day and 75 µg/day for women and men, respectively (see http://www.food.gov.uk/multimedia/pdfs/selenium.pdf), which is based on the intake required to support maximum expression of plasma glutathione peroxidase (GPx3). However, circulating concentrations of other plasma proteins, such as SePP, may reflect more accurately the requirements for Se (12)
. The need for a better definition of the Se requirements comes from observations that in many parts of the world, including Europe, mean Se intake is below the present recommendations, suggesting that a substantial proportion of people have a suboptimal Se status and therefore are at increased risk of disease (1
, 13)
. Such a view is supported by a study showing that 40% of elderly German women did not reach a desirable Se status (14)
and by observations that Se supplementation counteracts the effects of low intake on selenoprotein expression (12)
.
To assess Se requirements and the extent to which higher intakes may be desirable, it is necessary to identify factors that influence healthy Se status. One possible component is genetic factors that could give rise to interindividual variations in Se metabolism and response to Se supplementation. Several genetic variations in selenoprotein genes have been described previously (4)
, but there is little published information regarding their modulation of Se status and requirements. Of particular interest is the gene encoding SePP since this protein, which is the major selenoprotein in plasma, has a crucial role in Se transport and delivery of hepatic Se to other tissues (15
, 16)
. SePP is unique in containing multiple Sec per molecule (up to 10 in the human SePP). The generation of knock-out mouse models either deficient for the expression of SePP or conditionally deficient for the synthesis of tRNAsec (and consequently deficient for the expression of hepatic SePP) revealed an additional role of SePP as a specific Se supplier for the brain (17
18
19)
. Other studies demonstrate antioxidant properties of SePP (20
21
22)
. This bifunctionality of SePP may reflect the two regions of the human SePP that are apparent at the structural level, a N-terminal region with a single Sec and antioxidant properties and a C-terminal region containing nine Sec (23)
. Interestingly, Sec-containing fragments derived from the C terminus of SePP inhibited cell death of megakaryoblastoma (24)
.
The aims of the present work were 1) to identify common single nucleotide polymorphisms (SNPs) in the human SePP gene and 2) to assess their potential functionality by assessing their influence on markers of Se status and on the response of these markers to Se supplementation in healthy human volunteers (the SELGEN study). We identified two polymorphisms, one in the SePP coding region, confirming previous data (25)
, and one novel SNP in the region corresponding to the 3'UTR of SePP mRNA. Healthy volunteers were recruited prospectively based on their genotypes for each SNP and supplemented with 100 µg Se/day as sodium selenite for 6 wk followed by a 6 wk washout period. Measurement of several selenoprotein biomarkers of Se status revealed that both SNPs had functional effects on Se metabolism.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 27)
. We selected 40–100 blood samples at random from participants of European, Indian, and Chinese ethnic origins. DNA was extracted from frozen blood. Ten milliliters of blood were incubated in 4 vol. of reagent A (10 mM Tris HCl, pH8, 320 mM sucrose, 5 mM MgCl2, and 1% Triton-X100) for 5 min at room temperature and centrifuged for 4 min at 1300 g. The pellet was resuspended in 2 ml of reagent B (400 mM Tris HCl, pH8, 60 mM EDTA, 150 mM NaCl, and 1% SDS) and 500 µl 5M sodium perchlorate for deproteinization. Samples were briefly vortexed and incubated at room temperature for 15 min, followed by 30 min at 65°C; 2.5 ml of chloroform were added, and samples were incubated with rotation for 10 min at room temperature before being centrifuged at 250 g for 10 min at 4°C. DNA was precipitated in 2 vol. of 100% ethanol, and the DNA pellet was resuspended in 500 µl sterile water overnight at 4°C.
Polymerase chain reaction and DNA-high-performance liquid chromatography for SNP screening
DNA fragments from within the coding and 3'UTR regions of the SePP gene were amplified using primers and corresponding annealing temperatures presented in Table 1
. Positions of the start and end of the PCR fragments are indicated on the contig sequence containing the SEPP gene (NT_006576) as well as the size of polymerase chain reaction (PCR) products and amplified regions of the SEPP gene. PCR was performed with 100 ng template genomic DNA using Expand High Fidelity PCR system (Roche, Indianapolis, IN, USA) in a ThermoHybaid Px2 thermocycler with the following conditions: an initial denaturating step at 94°C for 4 min, followed by 30 cycles of denaturation (94°C for 30 s), annealing (see Table 1
for specific T°C for each PCR fragment, for 30 s), and extension (72°C for 1 min). PCR products were subjected to electrophoresis in a 1x Tris-Acetate-EDTA, 2% agarose gel and visualized with ethidium bromide.
|
Products were screened for heterozygosity using mutation detection (MD) denaturing high-performance liquid chromatography (dHPLC) (Transgenomic, Omaha, NE, USA). This technique relies on different thermodynamic properties between homoduplex and heteroduplex DNAs. Formation of homo- and heteroduplexes was performed by denaturation (95°C for 5 min) followed by slowly cooling to 40°C (0.03°C/second). DNA duplexes were then kept at 40°C for 5 min. Separation of the two forms of duplex DNAs by MD-dHPLC depends on the thermal stability of the duplex DNA, as determined by the melting temperature TM(50). Melting profiles were determined for each PCR product using Navigator software (Transgenomic) to calculate the 2–5 optimal temperatures for mutation detection analysis. Samples were injected at these different temperatures for separation by coupling ion-pair reverse phase liquid chromatography on a DNASep column in the presence of a bridging molecule [triethylammonium acetate (TEAA)]. DNA duplexes were eluted from the column by a linear acetonitrile gradient, and elution peaks were measured by an ultraviolet detector. PCR products showing a heterozygosity profile were sequenced to confirm the presence of polymorphism.
Se supplementation trial (the SELGEN study)
The study protocol was approved by the Sunderland Local Research Ethical Committee, and signed consent was given by all individuals. All study participants were recruited through poster advertisements from the general population and were residents in and around Newcastle-upon-Tyne (UK). The study included 121 unrelated healthy volunteers (male and female) aged 20–60 yr; all were nonsmokers. A questionnaire was used to assess general health and included questions on family history of cancer and cardiovascular problems and tobacco and alcohol consumption. Exclusion criteria included cardiovascular, hepatic, gastrointestinal, or thyroid disorders and cancer, as well as excessive alcohol consumption (>30 U/wk) and chronic intake of antiinflammatory drugs. People already taking selenium, multivitamins, or vitamin E supplements were excluded from the study. Weight and height were measured, and body mass index (BMI) was calculated. Peripheral blood samples were withdrawn from the antecubital vein into 10 ml EDTA-tubes (BD Vacutainer) and processed on the same day within 6 h of sample collection. All blood samples were taken in the morning after breakfast.
An initial 10 ml-blood sample was collected for assessing the genotype of each participant for two SNPs: a G/A variant at position 24731 in SePP mRNA, resulting in a amino acid change Ala234Thr (rs3877899) and a second G/A variant (rs7579) at position 25191 of the reference mRNA sequence NM_005410 in the 3'UTR of the mRNA. Seventy-five volunteers with appropriate genotype were asked to give a further 30 ml blood sample and then to take a daily supplement of 100 µg Se as sodium selenite for 6 wk. At the end of the supplementation period, another 30 ml blood sample was taken followed by three other blood samples at 2 wk intervals during a 6 wk washout period. Compliance with the selenium supplementation was estimated by counting the number of supplement capsules returned after the trial.
DNA extraction from fresh blood and genotyping
Blood samples were collected in 10 ml EDTA tubes and after separation of blood cells from plasma DNA extracted from the buffy coat using the Qiagen QIAmp DNA blood mini kit according to the manufacturer’s specifications. Both SNPs are present on the same 722bp-PCR product, and SePP genotype was determined by direct sequencing of PCR-amplification products. PCR was performed as described above using forward (cacgcattattcctatctctataagcttg) and reverse (ggaaatgaaattgtgtctagactaaattgg) primers as described in Table 1
. PCR products were subsequently PEG purified and sent for sequencing (MWG-Biotech, London, UK).
Plasma and erythrocyte preparation and isolation of lymphocyte-enriched fraction
After fractionation of the blood by centrifugation (4°C, 15 min, 950 g), plasma was further centrifuged for 12 min at 4°C, 730 g, to obtain platelet-free plasma. Erythrocyte fractions were immediately frozen. The lymphocytes were reconstituted in the same volume of RPMI1640/HEPES medium (Gibco-BRL, Gaithersburg, MD, USA) before being layered onto a 15 ml Histopaque H1077 gradient (Sigma-Aldrich, St. Louis, MO, USA) and centrifuged for 30 min at 4°C, 340 g (with no brake) to isolate the lymphocyte and granulocyte fractions. Cells were harvested, resuspended in 40 ml RPMI1640/HEPES, and pelleted by centrifugation (5 min, 4°C, at 340 g). At this stage, lymphocyte and granulocyte pellets were resuspended into 6 ml of RPMI1640/HEPES, aliquoted for protein or RNA preparation, and centrifuged for 15 min at 340 g, 4°C. Pellets designated for protein analysis were washed once in cold PBS and centrifuged for 10 min at 4°C at 13,800 g, and pellets were stored at –80°C until assays were carried out. Cell pellets designated for RNA extraction were resuspended in Trizol and frozen at –80°C.
Measurement of selenoprotein markers and plasma Se concentration
Lymphocyte GPx1 and plasma GPx3 activity were determined by the method of Paglia and Valentine (28)
as modified by Brown et al. (29)
using hydrogen peroxide as a substrate. One unit of GPx1 or GPx3 activity is defined as that which oxidizes 1 µmol of NADPH/min. Lymphocyte GPx4 activity was measured by the method of Weitzel et al. (30)
using phosphatidyl choline hydroperoxide as a substrate. One unit of GPx4 activity is defined as that which oxidizes 1 µmol of NADPH/min. Lymphocyte GPx1 and GPx4 protein levels were measured by competitive ELISA assay using polyclonal antibodies and recombinant proteins GPx1 and GPx4 obtained from Labfrontier (Seoul, Korea). The recombinant proteins were biotinylated using a commercial kit (Sigma). Standard curves of between 0.5 and 64 ng of protein for each assay were generated using recombinant GPx4 or GPx1 purified from human erythrocytes (Sigma). Sixty-five microliters of sample or standard were mixed with 65 µl of either biotinylated GPx1 or biotinylated GPx4 in round bottomed 96-well plates and left overnight at 4°C. High-binding, 96-well plates (Greiner, Stonehouse, UK) were coated with 100 µl antibody diluted in PBS (1:20,000 for GPx1 and 1:10,000 for GPx4) and incubated for 2 h at 37°C before blocking overnight at room temperature with 0.5% casein in PBS. After being washed three times with PBS containing 0.05% Tween, 100 µl of sample or standard were transferred to the respective high-binding plate and incubated at 37°C for 2 h, washed three times with PBS/Tween, and incubated with 100 µl NeutrAvidin HRP (Pierce, Rockford, IL, USA; diluted 1:10,000 in 0.5% casein) at 37°C for 1 h. After three washes with PBS/Tween, 100 µl TMB (KPL, Gaithersburg, MD, USA) were added to each well and incubated at room temperature for 15 min. The reaction was stopped with 100 µl 0.18 M H2SO4, and absorbance was determined at 450 nM. Plasma and erythrocyte TR1 concentrations were measured by radioimmunoassay as described previously (31)
. Calibrators were prepared from purified placental TR1. SePP in plasma was measured by radioimmunoassay as fully described by Burk et al. (32)
. Aliquots of a single purified human SePP were used to construct a calibration curve, and samples were analyzed together with a US reference plasma run at the same time so as to ensure the accuracy of the assay. Total plasma Se was measured by inductively coupled plasma mass spectrometry as described previously (33)
.
For all the above assays, samples were randomized across assays to avoid batch specific artifact. The analyst was unaware of the genotype and treatment of the samples as they were processed.
Statistics
The interdependence in the occurrence of different genotypes was assessed by a log-linear analysis of the frequencies of the different combinations. Biochemical variables were analyzed using factorial analysis of variance, with terms for gender, age, BMI, the three SNPs, and their interactions. When significant interaction between two factors was observed by ANOVA, secondary analysis was carried out to determine the main effects of each factor within subgroups. Data analysis was carried out on postsupplementation values adjusted for presupplementation values, and washout values were adjusted for postsupplementation values. The responses to Se supplementation and withdrawal were analyzed using a paired t test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and regions of the SePP gene corresponding to the protein coding and untranslated regulatory regions were amplified by PCR. These regions were screened for SNPs using dHPLC to identify regions of heterozygosity, followed by direct DNA sequencing to confirm presence of SNPs. Screening was carried out in 20 individuals and sequencing in 40–100, so only relatively common SNPs (present in >5%) would be detected. This approach allowed us to identify 2 G to A SNPs in the SePP gene. The first SNP, in the coding region at position 24731 of the mRNA, corresponds to a variant reported previously (25)
and results in an amino acid change (Ala to Thr) at position 234 of the SePP precursor, referred to in this article as Ala234Thr. The second SNP is novel, occurs in the 3'UTR at position 25191, and is described as r25191g/a. Table 2
shows the frequency of both polymorphisms in three major ethnic groups (Caucasian, Chinese, and South Asian) in the UK population. Both SNPs were observed in the three ethnic groups, and in all ethnic groups the frequency of the AA genotype was rare compared with the GG homozygote or GA heterozygote. Moreover, it was noted that homozygotes of AA genotype for one polymorphism were systematically GG for the other SNP (i.e., within our sample population, homozygotes AA for Ala234Thr SNP were always GG for r25191g/a SNP and homozygotes AA for r25191g/a were GG for Ala234Thr) indicating that, at least in this study, the two SNPs were not independent. The genotype distribution for both SNPs was in Hardy-Weinberg equilibrium. Haplotype analysis, using Haploview software, showed some evidence of linkage disequilibrium (D'=1 [confidence interval 0.63–1], r2=0.173).
|
An Se supplementation trial to assess functionality of the SNPs (the SELGEN study)
To study the influence of these SNPs on Se metabolism and/or response to Se supplementation, an intervention study was carried out in which markers of Se status were measured before and after a supplementation with 100 µg Se/day for 6 wk followed by a washout period of 6 wk (Fig. 1
A). Table 3
shows the characteristics of the 75 volunteers who were recruited and genotyped for both SNPs in SePP. The total cohort of 75 consisted of 40 individuals who were GG, 30 GA, and 5AA for SNP Ala234Thr and 37 who were GG, 32 GA, and 6AA for SNP r25191g/a.
|
|
Plasma Se concentrations were measured by inductively coupled plasma mass spectrometry (ICP-MS). Baseline (presupplementation) mean plasma Se for the whole cohort was 1.15 umol/l (±0.016) and this was raised (P<0.0001) by Se supplementation (100 µg/day) to 1.38 umol/l (±0.019; Fig. 1B
). Se withdrawal resulted in a rapid decrease in plasma Se status (P<0.0001) to baseline levels: plasma Se fell by 2 wk and continued to fall at 4 wk after which time it was similar to baseline concentration.
To study the functionality of the polymorphisms in SePP on plasma, erythrocyte and lymphocyte markers of Se status were measured and the data were analyzed depending on genotype for the SNPs in SePP, BMI, gender, and age. Additionally, the cohort was divided into tertiles depending on their baseline presupplementation plasma Se level of bottom Se Status (0.89–1.08 µmol/l), middle Se status (1.09–1.2 µmol/l), and highest Se status (1.21–1.55 µmol/l) to determine whether any effect of genotype was modulated by plasma Se concentration. While Se supplementation raised the plasma Se to a similar high level in all three tertiles, during the washout period the return of plasma Se to original baseline concentrations was significantly different for the three tertiles (P=0.002; Fig. 2
A).
|
As shown in Fig. 2B
, BMI affected baseline plasma Se concentration significantly (P=0.01), with plasma Se concentrations being lower in individuals who had a BMI of 25 or above (i.e., those who were overweight or obese). The association of BMI with Se status was abolished by supplementation suggesting that obese people may have a higher need for Se intake that is satisfied by Se supplementation. After withdrawal of the Se supplement, the fall in plasma Se was also influenced by BMI. When BMI was above 30, the fall was slower, with the difference from the postsupplementation level becoming significant only at 4 wk (P=0.016). Plasma Se concentration returned to baseline level in each BMI group after withdrawal of the Se supplement.
Effects of SNP Ala234Thr and SNP r25191g/a on plasma Se levels
The genotype for SNP Ala234Thr affected plasma Se concentration postsupplementation (P=0.012) and 2 wk after Se withdrawal (P=0.001; Fig. 2C
). Moreover, a significant interaction between genotype for SNP Ala234Thr and BMI was observed after supplementation (P=0.026) and after a 2 wk washout (P=0.008), with GG individuals of BMI > 30 having a higher plasma Se status compared with the other groups at the end of the supplementation period (Fig. 2C
). After a 2 wk washout, no decrease in plasma Se was observed in GG individuals of BMI > 30 compared with the other groups for which plasma Se had returned to baseline values. These data suggest that the genotype for SNPAla234Thr affects the response to supplementation in individuals with higher BMI.
Similar results were observed when plasma Se concentrations were analyzed according to SNP r25191g/a genotype (Fig. 2D
). A significant effect of genotype on plasma Se concentration was observed at the end of the supplementation period (P=0.002) and at 2 wk of washout (P=0.005). In addition, BMI interacted significantly with the genotype for SNP r25191g/a after supplementation (P=0.002) and 2 wk after Se withdrawal (P<0.001). In the presence of an A allele (GA+ AA), individuals of a BMI > 30 had higher concentration of plasma Se in response to Se (Fig. 2D
). No significant effects of BMI or genotypes in SePP were detectable at baseline or 4 wk and 6 wk after withdrawal of supplementation. Further analysis showed that SNP r25191g/a affected plasma Se at the end of the supplementation period (P=0.007) and after a 2 wk washout (P=0.047), with a higher plasma Se concentrations in homozygotes of AA genotype, irrespectively of BMI. After 4 wk Se withdrawal, SNP Ala234Thr and SNP r25191g/a interacted (P=0.046) in determining plasma Se status. Moreover, the fall in plasma Se concentrations occurred within 2 wk for all groups apart from GG individuals of BMI 25–30 for which Se fell progressively for 4 wk (paired t test, P<0.025) and (GA+AA) individuals with BMI > 30 for which the drop was dramatic and occurred at 4 wk. Additionally analysis of the data according to haplotype for both SNPs revealed an interaction between haplotype and BMI at the end of the supplementation (P<0.001) and 2 wk after withdrawal of Se (P<0.001). In summary, the data indicate that BMI interacts significantly with both SNPs, as well as with haplotype, to influence plasma Se status.
Effects of SNP Ala234Thr and SNP r25191g/a on plasma SePP
There was a highly significant (P<0.0001) 27% rise in mean plasma SePP level after Se supplementation, showing that plasma SePP concentration is a very sensitive biomarker of Se status, as has been indicated previously (12)
. Genotype for SNP Ala234Thr interacted significantly (P=0.028) with gender at baseline. As shown in Fig. 3
A, presupplementation SePP concentration was lower in female GA compared with GA males and higher in GG female compared with GG males. In contrast, SNP r25191g/a did not affect baseline SePP concentrations but did affect the concentration postsupplementation (genotypexgender interaction, P=0.018), with GA males having a higher plasma SePP concentration compared with GA females (Fig. 3B
). These data indicate that both SNPs may have functional effects on SePP expression. Interestingly, BMI had a significant effect at the end of the supplementation period (P=0.044), and genotype for SNP r25191g/a interacts significantly (P=0.040) with BMI (Fig. 3C
). Volunteers with a BMI < 25 had lower levels of SePP postsupplementation in GG than GA individuals. The effects of BMI are consistent with, but not as statistically significant as, the effect on plasma Se.
|
Effects of SNP Ala234Thr and SNP r25191g/a on plasma GPx3 activity and TR protein concentration
At baseline, gender significantly affected GPx3 activity (P=0.005) and interacted with BMI (genderxBMI; P=0.029), with lower GPx3 activity in obese women compared with obese males (Fig. 4
B). Se supplementation abolished this relationship, which was not restored even after the 6 wk washout period. However, it should be noted that there were only three obese women in this group. As shown in Fig. 4A
, dividing the volunteers into tertiles on the basis of plasma Se showed that baseline GPx3 increased with plasma Se concentration (P<0.0001). Although this effect on GPx3 disappeared after Se supplementation, it returned 6 wk after Se withdrawal (P=0.016). A significant increase (paired t test, P<0.0001) in GPx3 activity occurred in response to supplementation and a significant fall (paired t test, P<0.001) immediately after Se withdrawal. Further analysis showed that plasma GPx3 activity was lower in the age group (40–49 yr old) compared with other groups after supplementation (P=0.039) and during the washout period at all time points (all P<0.05). At baseline, despite a similar trend, the effect of age was not significant (data not shown). This could reflect different baseline dietary Se intake or age-related hormonal changes that affect GPx3 activity.
|
In individuals of GG genotype (but not GA) for SNP r25191g/a GPx3 activity increased significantly (paired t test, P=0.001) in response to Se supplementation with the difference in response between GG and GA genotypes approaching significance (ANOVA, P=0.087; Fig. 4C
). During Se withdrawal, GPx3 activity decreased overall (Fig. 4A
) but there were no consistent effects of genotype (Fig. 4C
). Moreover, genotype for SNP r25191g/a interacted significantly with gender at baseline (P=0.029), at the end of the supplementation (P=0.041) and 2 wk after Se withdrawal (P=0.012), with lower GPx3 activity in AA males. At baseline, significant effects of gender (P=0.038) and BMI (P=0.005) were observed.
Plasma TR1 concentration before supplementation was lower in AA homozygotes for SNP Ala234Thr (Fig. 5
A; P = 0.047). In addition, the ANOVA showed a significant interaction of genotype for SNP r25191g/a with gender (all P<0.033), observed at all time points apart form 4 wk after Se withdrawal. At baseline, males who were GG genotype for SNP r25191g/a had higher plasma TR1 concentrations than GG females and this effect disappeared after Se supplementation. Moreover, males of GA genotype for SNP r25191g/a were the only group to respond significantly to supplementation (paired t test, P=0.005; Fig. 5B
). ANOVA revealed that plasma TR1 levels were significantly different depending on BMI at 4 and 6 wk of the washout period (P
0.016), but there was no statistically significant difference after supplementation. Indeed TR1 levels were lower in lean (BMI<25) individuals than in obese individuals (BMI>30). Thus, both SNPs influenced plasma TR1 concentration and these effects were abolished by supplementation.
|
Effects of SNP Ala234Thr, SNP r25191g/a, gender on blood cell selenoprotein markers
Erythrocyte TR1 concentration was affected by SNP Ala234Thr (P=0.014), and the SNP was observed to interact significantly with gender (all P0.02) at all time points apart from 6 wk after Se withdrawal (Fig. 5C
). TR1 concentrations before supplementation were considerably lower in women of GA genotype compared with GG women, and this difference was maintained after Se supplementation and during the washout period. There was no significant effect in males, and thus it is possible that the effect of SNP Ala234Thr on erythrocyte TR1 level depends on gender. Likewise the ANOVA revealed an interaction of BMI and SNP Ala234Thr at all time points (all P<0.05), with higher levels of TR1 in erythrocytes of obese volunteers compared with lean individuals at all the time points. However, Se supplementation had no effect on erythrocyte TR1 levels. This could result from the fact that the life span of erythrocytes is longer than the 6 wk supplementation period.
Lymphocyte GPx1 and GPx4 activities and protein concentrations were determined by spectrophotometric assays and ELISA, respectively. Overall Se supplementation resulted in a small increase in GPx1 activity followed by a slow decline during the washout period (Fig. 6
A). This was accompanied by a small increase in protein level in response to the supplement followed by a biphasic fall during the washout (Fig. 6B
). Paired t test analysis revealed that the increase in GPx1 activity was significant in females of GA genotype for SNP Ala234Thr (P<0.0001) and GG males (P=0.008), for which this was accompanied by a significant increase in GPx1 protein concentration (P=0.011; Fig. 6A, B
). In addition, after only 2 wk Se withdrawal, GPx1 protein levels decreased in both GA females (P=0.028) and GG males (P=0.031). Moreover, GG females showed a rapid drop in GPx1 activity that was already significant after 2 wk withdrawal (P=0.029) with the interaction between gender and genotype on the decrease almost reaching significance (P=0.075). Interestingly, GA males did not show a significant increase or decrease in GPx1 activity or protein levels.
|
Likewise, SNP r25191g/a influenced the response of GPx1 activity to Se supplementation and activity was significantly affected by gender (P=0.037; Fig. 6C
). As analyzed by paired t test, there were significant increases in GPx1 activity in GG males (P=0.009), GA females (P=0.002), and GG females (P=0.019) but not in GA males (who already had the highest GPx1 concentrations), with a significant interaction between gender and SNP r25191g/a on the response of GPx1 (P=0.028). However, this effect on activity was not accompanied by a significant increase in protein concentration. After Se supplementation was withdrawn, GPx1 activities decreased significantly in all groups apart from GA males. "GPx1 activity/protein" ratio response to supplementation was significantly different between males GG and GA for SNP r25191g/a (P=0.042; data not shown). In addition, the fall in "GPx1 activity/ protein" ratio during the washout period was affected by the combination of genotype at SNP Ala234Thr and SNP r25191g/a (all time points P<0.05). In addition, SNP Ala234Thr and SNP r25191g/a interacted to influence GPx1 activity during the Se withdrawal (all P<0.01). Overall, the data suggest that genotype for SNP r25191g/a is associated with the amount of active GPx1 protein.
Lymphocyte GPx4 activity was not increased by Se supplementation, and there was little change during the washout period (Fig. 6D
). However, there was a significant interaction of genotype for SNPAla234Thr with gender at baseline (P=0.019) and the interaction almost reached significance after 6 wk Se withdrawal (P=0.06). GPx4 activity was lower in females who were GA and males who were GG for SNP Ala234Thr at baseline and after a 6 wk washout when GPx4 activity returned to baseline levels. In contrast, in males there was a trend toward lower GPx4 activity in individuals who were GG compared with those who were GA for SNP Ala234Thr. Furthermore, females of GG genotype tended to have lower GPx4 activity after supplementation, whereas females who were GA and AA showed a slight increase in activity. Moreover, GG females exhibited a significant decrease (paired t test, P<0.027) of GPx4 activity 2 wk after Se withdrawal. There were no effects of genotype on GPx4 protein concentration, suggesting that the SNP is associated with the level of active GPx4 protein.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, is present within the protein coding region of the gene and is predicted to cause an amino acid change (Ala to Thr). The second SNP, r25191g/a, is novel and located in a region corresponding to the 3'UTR and thus potentially could affect Sec incorporation during protein synthesis. Since in our population the two SNPs described represent the only two common variations within the protein coding region and 3'UTR of the SePP gene, it is likely that they contribute to the majority of interindividual genetic variation in SePP.
Supplementation with sodium selenite (providing 100 µg/day Se) for 6 wk increased plasma SePP levels by 27%. The relatively large increase in SePP concentration and the relatively small interindividual variation in response indicated that plasma SePP was more sensitive as a marker of Se status than either plasma Se or plasma GPx3, as described in an earlier study with a Chinese population (12)
. Interestingly, after supplementation the mean plasma SePP level was very close to the level of plasma SePP found in Americans on an unsupplemented diet (32)
. Samples from the SELGEN study were analyzed in parallel with a US reference sample, and therefore the values obtained for the UK and US samples are comparable. It is well established that Se status in the UK is below that in the US (34)
. The data indicate that daily supplementation with 100 µg Se as sodium selenite is sufficient to bring UK Se status, as assayed by plasma SePP, close to an American level. Genotype analysis suggested that both SNPs influenced SePP levels. SNP Ala234Thr affected baseline SePP while in contrast SNP r25191g/a affected the level postsupplementation. There is a potential mechanistic basis for the two SNPs having quite different effects: SNP Ala234Thr is in the protein coding region and causes an amino acid change, whereas SNP r25191g/a is within the 3'UTR and may therefore affect Sec incorporation and SePP synthesis. Our hypothesis is that SNP Ala234Thr affects the stability of SePP, possibly through some post-translational modifications, which in turn could affect protein levels only when Se intake is suboptimal (i.e., as seen with UK baseline levels). On the other hand, the SNP r25191g/a could affect the efficiency of SePP synthesis in response to increased intake in Se by inducing a change in the SECIS function. Statistical analysis indicated that the two SNPs are linked, and it may be that the combination of the two SNPs leads to either lower stability with higher synthesis or vice versa.
There is now strong evidence that SePP plays a critical role in delivery of hepatic Se to other tissues and knock-out of the gene affects Se transport (15
, 16)
. For this reason, more subtle alterations in SePP concentration or function would be expected to change Se supply to different tissues and therefore expression of other selenoproteins. Indeed, this view was supported by the observations that five biomarkers of Se status and metabolism were influenced by genotype at these two loci within the SePP gene. First, plasma Se levels were linked to BMI to a greater extent in GG individuals for SNP Ala234Thr than in the GA heterozygotes (Fig. 2)
. Second, plasma TR1 concentration was influenced by both SNP Ala234Thr and SNP r25191g/a at baseline but not after Se supplementation for 6 wk. Third, erythrocyte TR1 concentration was higher in females who were GG for SNP Ala234Thr than in the GA heterozygotes. Fourth, lymphocyte GPx1 protein and activity levels responded to Se supplementation to a greater extent in GG individuals than in the heterozygotes in the case of both SNPs. Finally, baseline lymphocyte GPx4 activity was different in individuals who were GG compared with GA for SNP Ala234Thr. Overall the data indicate that the two SNPs have functional effects and that genetic variation in the SePP gene is associated with differences in selenoprotein metabolism.
In addition, statistical analyses indicate that several effects of the two SNPs are influenced by gender and BMI, emphasizing the complex nature of the interactions with Se metabolism. For example, there may be an effect of steroid hormones, dependent on the sex of the individual as well as hormones and cytokines originating from adipose tissue. Interestingly, gender-specific differences in selenoprotein expression occur in rodents, with females exhibiting higher serum GPx3 activity and higher liver SePP and deiodinase mRNA levels (35
, 36)
. Little information is available on the relationship between BMI and Se status. No interaction between serum Se concentration and BMI was found in elderly German women (14)
. In contrast, there was a negative correlation between BMI and Se status in the EVA study (10)
. In the present SELGEN study, there was an association of BMI with various selenoprotein biomarkers (e.g., plasma Se, GPx3, and SePP) and these effects of BMI were influenced by genetic variations in SePP. Moreover the association between BMI and Se status was abolished by Se supplementation, suggesting that these effects of BMI only occur at suboptimal Se intake. From a public health point of view, it suggests that, compared with people of normal weight, overweight and obese people in the UK may have 1) higher Se needs, 2) lower Se intakes, or 3) both so that their usual Se status is relatively low and that they may benefit even more from raised Se intakes. The observations from the present study will need to be confirmed in larger studies in which subjects are recruited prospectively on the basis of their BMI. The mechanism(s) for the association of BMI with Se metabolism are not known, but they could reflect links the between BMI and thyroid hormone metabolism and the role of Se in the thyroid (2)
.
In conclusion, this study demonstrates that daily supplementation of healthy volunteers in the UK with 100 µg Se (as sodium selenite) increases Se status to the levels typical of the USA population, as judged by plasma SePP concentrations. Furthermore, both baseline and postsupplementation levels of selenoprotein markers are influenced by genetic factors. The results of the Se supplementation trial provide the first evidence that these variants affect Se metabolism and the response of selenoprotein metabolism to supplementation; thus there may be an important genetic component in influencing interindividual differences in Se metabolism. However, further biochemical studies are required to link directly the two genetic variants described here to functional changes in SePP. Crucially the biological outcomes of the two SNPs are superimposed on those of gender and BMI, thus emphasizing how these various factors interact to influence Se metabolism and so help predict their potential impact on Se intake and public health.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 7, 2007. Accepted for publication April 26, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Meplan, L. K Crosley, F. Nicol, G. W Horgan, J. C Mathers, J. R Arthur, and J. E Hesketh Functional effects of a common single-nucleotide polymorphism (GPX4c718t) in the glutathione peroxidase 4 gene: interaction with sex Am. J. Clinical Nutrition, April 1, 2008; 87(4): 1019 - 1027. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Pagmantidis, C. Meplan, E. M van Schothorst, J. Keijer, and J. E Hesketh Supplementation of healthy volunteers with nutritionally relevant amounts of selenium increases the expression of lymphocyte protein biosynthesis genes Am. J. Clinical Nutrition, January 1, 2008; 87(1): 181 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Patsopoulos, A. Tatsioni, and J. P. A. Ioannidis Claims of Sex Differences: An Empirical Assessment in Genetic Associations JAMA, August 22, 2007; 298(8): 880 - 893. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
). B) Histogram showing mean ± SE of plasma Se (µmol/l). Statistical analysis was carried out on adjusted values as described in Materials and Methods. Plasma Se rose significantly on supplementation and fell within 2 wk during washout period (***P<0.001). Presupp = presupplementation; postsupp = postsupplementation.
