Advertisement

Genes & Nutrition

, 13:20 | Cite as

Genetic polymorphism in selenoprotein P modifies the response to selenium-rich foods on blood levels of selenium and selenoprotein P in a randomized dietary intervention study in Danes

  • Tine Iskov Kopp
  • Malene Outzen
  • Anja Olsen
  • Ulla Vogel
  • Gitte Ravn-Haren
Open Access
Research

Abstract

Background

Selenium is an essential trace element and is suggested to play a role in the etiology of a number of chronic diseases. Genetic variation in genes encoding selenoproteins, such as selenoprotein P and the glutathione peroxidases, may affect selenium status and, thus, individual susceptibility to some chronic diseases. In the present study, we aimed to (1) investigate the effect of mussel and fish intake on glutathione peroxidase enzyme activity and (2) examine whether single nucleotide polymorphisms in the GPX1, GPX4, and SELENOP genes modify the effect of mussel and fish intake for 26 weeks on whole blood selenium, plasma selenoprotein P concentrations, and erythrocyte GPX enzyme activity in a randomized intervention trial in Denmark.

Results

CC homozygotes of the SELENOP/rs3877899 polymorphism who consumed 1000 g fish and mussels per week for 26 consecutive weeks had higher levels of both selenoprotein P (difference between means − 4.68 ng/mL (95% CI − 8.49, − 0.871)) and whole blood selenium (difference between means − 5.76 (95% CI − 12.5, 1.01)) compared to fish and mussel consuming T-allele carriers although the effect in whole blood selenium concentration was not statistically significant.

Conclusions

Our study indicates that genetically determined variation in SELENOP leads to different responses in expression of selenoproteins following consumption of selenium-rich foods. This study also emphasizes the importance of taking individual aspects such as genotypes into consideration when assessing risk in public health recommendations.

Keywords

Selenium Selenoprotein P Glutathione peroxidase Dietary intervention Randomized Gene-diet interaction SELENOP GPX1 GPX4 Single nucleotide polymorphisms 

Abbreviations

BMI

Body mass index

GPX

Glutathione peroxidases

SNP

Single nucleotide polymorphism

Background

Selenium is an essential trace element and is suggested to play a role in the etiology of a number of chronic diseases [9, 37]. Several factors can affect selenium status; these include exposure to selenium through diet or dietary supplements [32, 33, 35] and lifestyle factors such as body mass index (BMI) and smoking [6]. Furthermore, genetic variation has been proposed to influence selenium status of the individual [14].

Most of the biological functions of selenium are carried out by selenoproteins [30]. Among the best characterized selenoproteins with essential functions are selenoprotein P and the glutathione peroxidases (GPX), including GPX1 and GPX4 [30]. Besides functioning as storage and transport protein for selenium, some studies suggest that selenoprotein P is involved in the antioxidant defense which also comprises the glutathione peroxidases [40]. These antioxidant selenoenzymes detoxify a range of hydroperoxides, including lipid and phospholipid hydroperoxides, thereby ensuring a reducing environment and protecting cell components such as proteins, lipids, and DNA from oxidative damage [36]. Nutritional deficiency in selenium results in decreased selenoprotein concentrations and a compromised enzymatic antioxidant defense. Selenium supplementation is known to affect selenoprotein expression in a hierarchical manner. GPX4 ranks higher in the hierarchy of selenoproteins compared to GPX1 which is much more responsive to both selenium depletion and repletion [2, 4]. Humans differ in their ability to metabolize selenium and respond differently to selenium supplementation. These differences are likely to be due to genetic variation in selenoprotein genes [6, 16]. Specifically, functional single nucleotide polymorphisms (SNPs) in GPX1 (encoding GPX1), GPX4 (encoding GPX4), and SELENOP (encoding SELENOP) have been shown to affect blood selenium or selenoprotein levels in response to supplementation [19, 20, 23]. The GPX1/rs1050450 polymorphism is a Pro to Leu substitution at position 197 [10] which results in reduced enzyme activity [16, 34] and higher DNA damage levels [25]. GPX1/rs1050450 has also been associated with risk of several diseases, including lung [31, 39], breast [8, 21, 34], and prostate cancer [3, 43]; peripheral neuropathy [45]; coronary heart diseases [26, 50]; septic shock [18]; and mortality [42], and correlation between GPX1 activity and serum selenium concentration has been reported for this polymorphism [16]. GPX4/rs713041 affects protein binding to the 3′-untranslated region close to a selenocysteine insertion sequence required for selenoprotein synthesis [20, 48] and has been associated with decreased GPX enzyme activity [21], risk of colorectal cancer [43], and mortality [47]. SELENOP/rs3877899 causes an Ala to Thr amino acid substitution at position 234, and SELENOP/rs7579 is located in the 3′-untranslated region of SELENOP mRNA, where it causes a G to A base change [19]. Both SELENOP polymorphisms lead to alterations in selenium metabolism by changing the proportion of the 60-kDa isoform of selenoprotein P [23]. They have been shown to affect selenium and selenoprotein P blood concentrations after supplementation with selenium [19] and have also been associated with risk of various diseases, including breast [21], prostate [7, 13, 43], and colorectal cancer [22, 23] and aortic aneurisms [44].Taken together, these results point to a regulatory role of these polymorphisms in the expression of the encoded selenoproteins. However, it is not clear whether this is also the case at selenium intakes that are dietary feasible.

In a previous randomized dietary intervention with fish and mussels in middle-aged Danes, we found increased whole blood selenium and plasma selenoprotein P concentrations among healthy participants after 26 weeks’ intervention [28]. We hypothesized that SELENOP gene variations modify the effect of increased dietary selenium intake on markers of selenium status. Thus, in the present paper, we aimed to (1) investigate the effect of mussel and fish intake on GPX enzyme activity and (2) examine whether SNPs in the GPX1, GPX4, and SELENOP genes modify the effect of the intervention with mussel and fish on whole blood selenium, plasma selenoprotein P concentrations, and erythrocyte GPX enzyme activity in the same study population of healthy middle-aged Danish men and women.

Methods

Study population

This study is based on data from a randomized dietary intervention study with the primary aim of studying the influence of increased intake of fish and mussels on blood selenium levels. The study design and methods are described in details elsewhere [28].

The study took place in the northern part of Jutland, Denmark, from September 2010 to March 2011 as a 26-week randomized dietary intervention study with two parallel groups including healthy middle-aged participants. Eligibility for randomization was determined by potential participants completing a questionnaire 3 months before baseline measurements on lifestyle, diet, and health status.

The goal was to recruit a total of 100 men and women aged 50–74 years with a BMI of 18.5–28 kg/m2 based on a power calculation where a minimum detectable difference of 10 ng/mL (SD = 10) or 30 ng/mL (SD = 10) in selenium concentration between groups with a statistical power of 87 or 99%, respectively, was allowed [29]. The exclusion criteria included current smoking, intake of dietary supplements containing selenium 3 months before baseline measurements, frequent intake of fish and shellfish (> 300 g/week), excessive intake of alcohol (according to the official Danish guidelines at study recruitment: women > 14 units of alcohol/week, men > 21 units of alcohol/week), strenuous exercise (> 10 h/week), severe chronic disease, and frequent use of specified medication (including diabetic medicine, anticoagulant medicine, and medication for heart disease), or a cancer diagnosis within the past 5 years. Study participants were requested to inform the investigators of any changes regarding disease or medication occurring during the study. Study participants were recruited through local media, including newspaper advertisements. Of the 102 men and women enrolled in the study, 83 completed the intervention. Flow chart of the participants is illustrated in Additional file 1 as previously published [28].

Dietary intervention

Participants in the intervention group were provided with 1000 g raw fish and raw or processed mussels (portion size of 200 g; five portions/week) once a week for 26 weeks. This amount corresponds to an intake of approximately 50.3 μg selenium/day (based on data from the Danish Food Composition Databank) [41]. The participants received four or five different types of fish per week. Diversity in the type of fish provided was prioritized to ensure variation and thereby optimize compliance. The participants were allowed to consume other meals containing fish or shellfish besides the experimental foods. The intervention diet has been described in detail elsewhere [28].

To monitor their compliance, the participants were provided with a self-monitoring record and kitchen scales to weigh the amount of received fish and mussels (prepared or raw) not consumed during the study period. At study initiation, participants were instructed on how to complete the self-monitoring record. Compliance was calculated as total received amount—not ingested amount after preparation)/by total received amount; the median of ingested proportion was 99% of the received fish and mussels [28].

Participants in the control group received no intervention and were advised to maintain their habitual diets.

Data collection

Non-fasting blood samples were collected three times (weeks 0, 13, and 26) from each participant during the study. Blood samples were drawn in K2-EDTA-coated blood drawing tubes as whole blood samples or blood samples that were separated into plasma, erythrocytes, and buffy coat by centrifugation. All samples were stored at − 80 °C until analysis.

Biochemical analyses

Whole blood selenium analyses were conducted using an ELAN 6100 DRC inductively coupled plasma-mass spectrometer (ICP-MS) in accordance with a method described in detail in [15]. The concentration of selenoprotein P in plasma was determined using its selective retention by heparin-affinity high-performance liquid chromatography (HPLC) and online detection by ICP-DRC-MS of selenium eluting from the HPLC column. In addition to determining selenoprotein P, the concentration of total selenium in plasma was also quantified by isotope dilution on the basis of the area under the complete chromatogram. However, even though the results seemed accurate (that is, the values obtained from plasma selenium did not deviate from certified values from reference material), the method used to determine plasma selenium in this way was less precise than that used to determine selenium in whole blood. Therefore, the baseline plasma selenium concentration was used only for comparison with other studies measuring plasma selenium in healthy populations. The analysis methods have been described in details previously [28].

GPX enzyme activities were spectrophotometrically assayed in erythrocyte lysates on a Pentra 400 (Horiba ABX, Montpellier, France), using t-butylhydroperoxide as substrate according to the method described by Wheeler et al. [49]. GPX activities were related to the amount of hemoglobin (Hb) in the lysates. Hb content was determined using a commercially available kit (Randox Ardmore, UK, cat. no HG 980). Samples from each individual were run in the same batch in random order to decrease variation, and a control sample was included for every 15th sample. Intra- and inter-day variations were < 5 and < 10%, respectively.

SNP selection and genotyping

The following polymorphisms were selected based on their known functionality and association with disease: GPX1/rs1050450, GPX4/rs713041, SELENOP/rs3877899, and SELENOP/rs7579.

DNA from the participants was extracted from frozen lymphocytes as described [24]. Genotypes were determined using RT-PCR and allelic discrimination on ABI 7900HT instruments (Applied Biosystems, Nærum, Denmark). Generally, 40–200 ng/μl DNA was obtained and 10 ng of DNA was genotyped in 5 μl containing 50% 2 × Mastermix (Applied Biosystems, Nærum, Denmark), 100 nM probes, and 900 nM primers.

GPX1/rs1050450: primers: F: 5′-TGT GCC CCC TAC GCA GGT ACA-3′, R: 5′-CCC CCG AGA CAG CAG CA-3′ (TAGCopenhagen, Copenhagen, Denmark), T-allele: 5′-FAM-CTG TCT CAA GGG CTC AGC TGT-MGB-3′, C-allele: 5′-VIC-CTG TCT CAA GGG CCC AGC TGT-MGB-3′ (Applied Biosystems, Nærum, Denmark).

GPX4/rs713041: primers: F: 5′-CCC ACT ATT TCT AGC TCC ACA AGT G-3′, R: 5′-GTC ATG AGT GCC GGT GGA A-3′ (TAGCopenhagen, Copenhagen, Denmark), T-allele: 5′-FAM-ACG CCC TTG GAG C-MGB-3′, C-allele: 5′-VIC-ACG CCC TCG GAG C-MGB-3′ (Applied Biosystems, Nærum, Denmark).

SELENOP/rs3877899 was determined using the TaqMan® Pre-designed SNP genotyping; assay ID C_2841533_10, respectively (Applied Biosystems, Nærum, Denmark).

SELENOP/rs7579: primers: F: 5′-CAA AAA AGT GAG AAT GAC CTT CAA ACT-3′, R: 5′-ATG CTG GAA ATG AAA TTG TGT CTA GA-3′ (TAGCopenhagen, Copenhagen, Denmark), G-allele: 5′-FAM-AAA ATA GGA CAT ACT CCC C-MGB-3′, A-allele: 5′-VIC-AAA TAG AAC ATA CTC CCC AAT T-MGB-3′ (Applied Biosystems, Nærum, Denmark).

All samples were determined as duplicates with known positive controls (three for each genotype) and three negative controls containing Milli Q water. All duplicates yielded 100% identical genotypes.

Statistical methods

The statistical analysis was based on all available observations. Participants who were randomized into a group, but did not attend the baseline appointment (week 0), were excluded. Baseline characteristics are presented as either number with its percentage and medians with 5th and 95th percentiles for each study group. For the polymorphisms, minor allele frequencies are also presented. Outlying observations were identified from visual inspection of the data (correlation plots of week 0 vs. week 13 and week 0 vs. week 26). Deviation from Hardy–Weinberg equilibrium was assessed using a chi-square test.

Linear multiple regression analysis was applied to evaluate the intervention effect on erythrocyte GPX enzyme activity. Within the two groups, mean changes (weeks 0–13, weeks 0–26) and the difference between the groups’ mean changes (weeks 0–26) were calculated from the linear multiple regression. All values of GPX enzyme activity were log-transformed to correct for right-skewed distribution.

To examine the association between mean concentrations of whole blood selenium, plasma selenoprotein P or erythrocyte GPX enzyme activity according to genotype at baseline, week 13 and week 26, respectively, a linear multiple regression analysis was applied using least square means adjusted for baseline concentrations of whole blood selenium, plasma selenoprotein P, or erythrocyte GPX enzyme activity, respectively. Adjustment for baseline level was done to eliminate baseline levels’ influence on the effect of the SNPs. In order to increase the statistical power, heterozygote variant allele and homozygote variant allele carriers were pooled in the analyses.

Moreover, we examined whether sex, BMI, and age modified the relationship between intervention outcomes and the polymorphisms. An interaction term between sex, BMI, and age, respectively, and the studied polymorphisms was therefore included in the model.

A mixed-model, repeated-measures analysis of variance (ANOVA) was used to determine the within-subject effect between genotype and erythrocyte GPX enzyme activity, whole blood selenium, or plasma selenoprotein P concentrations during the entire intervention period.

The statistical analyses were carried out using SAS (release 9.4, SAS Institute, Inc., Cary, NC, USA) and the procedure general linear model (GLM).

For all tests, a P value less than 0.05 was considered statistically significant.

Results

We evaluated the effect of four functional polymorphisms in GPX1, GPX4, and SELENOP on erythrocyte GPX enzyme activity, whole blood selenium, and plasma selenoprotein P concentrations after consumption of 1000 g fish and mussels per week for 26 consecutive weeks (~ 50 μg selenium/day) in volunteer participants [28]. Furthermore, we evaluated the effect of the intervention on erythrocyte GPX enzyme activity.

Baseline characteristics of all participants including the control group are presented in Table 1 as published previously [28], except for GPX enzyme activity measurements, which have not been published previously. None of the baseline variables differed between the two groups (Table 1) or according to genotype (results not shown). The genotype distributions of the studied polymorphisms were in Hardy–Weinberg equilibrium (results not shown).
Table 1

Baseline characteristics of the participants presented as either number (%) or median value (5th–95th percentiles)

Variable

MAF (%)

Intervention group (n = 49)

Control group (n = 45)

Women

 

21 (43%)

19 (42%)

Men

 

28 (57%)

26 (58%)

Age, years

 

61 (51–72)

59 (51–73)

BMI, kg/m2

 

26.3 (21.0–31.1)

25.2 (20.3–32.5)

Whole blood selenium, ng/mL

 

113.5 (91.3–147.4)

114.6 (96.6–136.0)1

Plasma selenoprotein P, ng selenium/mL

 

51.4 (35.0–63.9)2

51.4 (35.4–65.7)3

Plasma selenium, ng/mL

 

84.7 (67.8–106.5)2

86.4 (70.5–103.3)3

Erythrocyte GPX enzyme activity, U/g Hb

 

86 (57–153)4

84 (54–156)5

GPX1/rs1050450

29.8

  

CC

 

23 (47%)

22 (49%)

CT

 

22 (45%)

20 (44%)

TT

 

4 (8%)

3 (7%)

GPX4/rs713041

37.8

  

CC

 

16 (33%)

17 (38%)

CT

 

25 (51%)

26 (58%)

TT

 

8 (16%)

2 (4%)

SELENOP/rs3877899

19.1

  

CC

 

30 (61%)

30 (67%)

CT

 

18 (37%)

14 (31%)

TT

 

1 (2%)

1 (2%)

SELENOP/rs7579

30.3

  

GG

 

24 (49%)

19 (42%)

GA

 

22 (45%)

23 (51%)

AA

 

3 (6%)

3 (7%)

Part of this table has been published in [28]. MAF, minor allele frequency for the studied population (n = 94). 1n = 44 due to exclusion of outlying values (n = 1), all whole blood and plasma selenium and selenoprotein P concentrations for this person were excluded; 2n = 46 due to errors in the laboratory measures (n = 3), all plasma selenoprotein P concentrations and the plasma selenium concentration analyzed only at baseline were excluded for these persons; 3n = 42 due to errors in the laboratory measures (n = 2), all plasma selenoprotein P concentrations and the plasma selenium concentration analyzed only at baseline were excluded for these persons, and due to exclusion of outlying values (n = 1). 4n = 44 since GPX activity was not analyzed on persons having only baseline levels measured (n = 1). 5n = 41 since GPX activity was not analyzed on persons having only baseline levels measured (n = 8)

In Table 2, changes in erythrocyte GPX enzyme activity measurements within groups (intervention and control) and differences between group changes are illustrated. There were no statistically significant differences in erythrocyte GPX enzyme activity changes between the intervention and control groups after, neither 13 nor 26 weeks (Table 2).
Table 2

Changes within groups and differences between group changes in erythrocyte GPX enzyme activities

 

Mean change within group, log (U/g Hb) (95% CI)a

Difference between group mean change, log (U/g Hb) (95% CI)*

 

N

Weeks 0–13

n

Weeks 0–26

N

Weeks 0–26

Intervention

41

0.0332 (0.0160, 0.0503)

41

0.0461 (0.0239, 0.0684)

85

0.0263 (− 0.00498, 0.0577)

Control

44

0.0169 (0.000278, 0.0334)

44

0.0198 (− 0.00221, 0.0418)

aAdjusted for baseline levels of GPX enzyme activity

In Table 3, the associations between mean erythrocyte GPX enzyme activity, concentrations of whole blood selenium and plasma selenoprotein P, and the studied polymorphisms, and within-subject effects between genotype and time for the intervention group, are shown. A difference in mean GPX enzyme activity at baseline was found for the GPX1/rs1050450 polymorphism among participants who were randomized to the intervention (P = 0.044). This difference in enzyme activity persisted after 13 and 26 weeks’ intervention, resulting in no statistically significant difference in response to the increased selenium intake between genotypes. Mean GPX enzyme activity at baseline among CC homozygotes was 104.1 and 84.0 U/g Hb among variant T-allele carriers (intervention group only—results not shown). There was no interaction between any of the studied polymorphisms and whole blood selenium or plasma selenoprotein P concentrations (Table 3). However, there was a statistically significant difference of − 4.68 ng/mL (95% CI − 8.49, − 0.871) between mean concentration of plasma selenoprotein P at week 26 for variant T-allele and CC homozygotes of the SELENOP/rs3877899 polymorphism (Table 3). A mean difference in whole blood selenium for the SELENOP/rs3877899 polymorphism was also seen; however, this association was not statistically significant (difference between means − 5.76 (95% CI − 12.5, 1.01). Results for the control group are presented in Additional file 2. None of the polymorphisms among participants in the control group differed for either of the outcome measures as expected.
Table 3

Association between mean concentrations of erythrocyte GPX enzyme activity, whole blood selenium, and selenoprotein P in relation to the studied polymorphisms, and within-subject effects between genotype and time in the intervention group

Erythrocyte GPX enzyme activity, log (U/g Hb)

SNP

n

Baseline

13 weeksc

26 weeksc

P value for interaction between genotype and time

 

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

GPX1/rs1050450

 CC

20

− 0.224 (− 0.442, − 0.00638)

0.044

0.0299 (− 0.00981, 0.0696)

0.14

− 0.00585 (− 0.06278, 0.0511)

0.84

0.085

 CT + TT

21

GPX4/rs713041

 CC

15

0.129 (− 0.105, 0.364)

0.27

− 0.00532 (−  0.0462, 0.0355)

0.79

− 0.0143 (− 0.0711, 0.0424)

0.61

0.79

 CT + TT

26

SELENOP/rs3877899

 CC

25

− 0.0228 (− 0.258, 0.212)

0.85

− 0.00617 (− 0.0459, 0.0335)

0.75

0.0145 (− 0.0407, 0.0697)

0.60

0.70

 CT + TT

16

SELENOP/rs7579

 GG

22

− 0.0278 (− 0.258, 0.202)

0.81

− 0.0134 (− 0.0521, 0.0252)

0.49

− 0.0143 (− 0.0683, 0.0397)

0.60

0.81

 GA + AA

19

Whole blood selenium, ng/mL

SNP

n*

Baseline

13 weeksd

26 weeksd

P value for interaction between genotype and time

 

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

GPX1/rs1050450

 CC

23

3.20 (− 5.46, 11.9)

0.46

− 0.989 (−7.89, 5.91)

0.77

− 1.24 (− 8.11, 5.63)

0.72

0.83

 CT + TT

26

GPX4/rs713041

 CC

16

4.85 (− 4.32, 14.0)

0.29

− 2.76 (− 9.91, 4.38)

0.44

− 1.77 (− 8.93, 5.39)

0.62

0.54

 CT + TT

33

SELENOP/rs3877899

 CC

30

0.989 (− 7.93, 9.91)

0.82

0.675 (− 6.37, 7.72)

0.85

− 5.76 (− 12.5, 1.01)

0.093

0.13

 CT + TT

19

SELENOP/rs7579

 GG

24

4.44 (− 4.16, 13.0)

0.30

− 4.04 (− 10.9, 2.80)

0.24

1.60 (− 5.32, 8.53)

0.64

0.23

 GA + AA

25

Selenoprotein P, ng/mL

SNP

n b

Baseline

13 weekse

26 weekse

P value for interaction between genotype and time

 

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

Difference between means (95% CI)

P value

GPX1/rs1050450

 CC

21

4.17 (− 0.51, 8.85)

0.080

− 1.17 (− 5.38, 3.04)

0.58

− 2.56 (− 6.60, 1.47)

0.21

0.071

 CT + TT

25

GPX4/rs713041

 CC

16

3.49 (− 1.47, 8.45)

0.16

0.801 (− 3.41, 5.01)

0.70

− 1.17 (− 5.32, 2.97)

0.57

0.48

 CT + TT

30

SELENOP/rs3877899

 CC

28

− 2.33 (− 7.23, 2.57)

0.34

− 2.43 (− 6.53, 1.67)

0.24

− 4.68 (− 8.49, − 0.871)

0.018

0.33

 CT + TT

18

SELENOP/rs7579

 GG

23

2.16 (− 2.63, 6.94)

0.37

1.72 (− 2.35, 5.79)

0.40

− 0.630 (− 4.66, 3.40)

0.75

0.79

 GA + AA

23

a13 and 26 weeks’ measurements only included 41 participants due to discontinuation of intervention

b13 and 26 weeks’ measurements only included 38 participants due to discontinuation of intervention

cAdjusted for baseline levels of erythrocyte GPX enzyme activity

dAdjusted for baseline levels of whole blood selenium

eAdjusted for baseline levels of selenoprotein P

In order to elucidate whether these differences in selenoprotein P and whole blood selenium concentrations were due to chance or an effect of the intervention, we examined the association between the SELENOP/rs3877899 polymorphism and whole blood selenium and selenoprotein P concentrations, respectively, with the control group included in the model (Figs. 1 and 2). After 26 weeks, the difference in whole blood selenium mean concentrations between CC homozygotes and T-allele carriers was only borderline statistically significant (P = 0.088) (Fig. 1), whereas the mean difference in selenoprotein P concentrations between CC homozygotes and T-allele carriers of the SELENOP/rs3877899 polymorphism was still statistically significant (P = 0.048) (Fig. 2).
Fig. 1

Association between mean concentrations of whole blood selenium and the SELENOP/rs3877899 polymorphism. Mean concentrations of whole blood selenium ± SD for the intervention and control group estimated by linear multiple regression adjusted for baseline level of whole blood selenium. P values for difference in mean whole blood selenium concentrations at week 26 for genotype effect within the intervention group (*) and within the control group (**), respectively, are illustrated

Fig. 2

Association between mean concentrations of selenoprotein P and the SELENOP/rs3877899 polymorphism. Mean concentrations of selenoprotein P ± SD for the intervention and control group estimated by linear multiple regression adjusted for baseline level of selenoprotein P. P values for difference in mean selenoprotein P concentrations at week 26 for genotype effect within the intervention group (*) and within the control group (**), respectively, are illustrated

We also tested whether sex, age, or BMI modified the effect on the intervention for the studied polymorphisms, but we did not find any sign of such effect modification (results not shown).

Discussion

The present study showed that CC homozygotes of the SELENOP/rs3877899 polymorphism who consumed 1000 g fish and mussels per week for 26 consecutive weeks had higher levels of both selenoprotein P and whole blood selenium compared to fish and mussel consuming variant T-allele carriers and the control group although the effect in whole blood selenium concentration was not statistically significant.

To our knowledge, this is the first study investigating the effect of SELENOP and GPX polymorphisms after ingestion of high selenium content foods in a controlled randomized trial. The SELGEN study [19] examined the effect on plasma selenium and selenoprotein P concentrations of the same SELENOP polymorphisms as in the present study before and after selenium supplementation in a UK population also known to be low in selenium status. Mean baseline plasma selenium concentrations in the two studies were comparable (1.15 μmol/l corresponding to 90.8 ng/mL in the SELGEN study and 85.8 ng/mL in the present study). The two investigated SELENOP polymorphisms significantly affected plasma selenium concentration after 6 weeks of supplementation with 100 μg selenium/day as sodium selenite [19]. The authors noted that the effect was strongest among participants with a BMI exceeding 30 where SELENOP/rs3877899 CC homozygotes (referred to as GG in [19]) and variant A-allele carriers of SELENOP/rs7579 responded better to selenium supplementation compared to carriers of the variant T-allele (referred to as A-allele in [19]) and GG homozygotes, respectively. The same study reported a statistically significant increase in plasma selenoprotein P concentration following selenium supplementation and gender-specific differences in baseline concentrations and post-supplementation with SELENOP/rs3877899 and SELENOP/rs7579, respectively. Lower baseline concentration of plasma selenoprotein P was measured in women who were heterozygous T-allele carriers of the SELENOP/rs3877899 polymorphism compared to men with corresponding genotype, while the opposite applied for CC homozygotes. Among carriers of the SELENOP/rs7579 variant A-allele, men had higher plasma selenoprotein P concentrations post-supplementation compared to women. We were not able to reproduce the findings on gender and BMI which could be due to a lack of power to study gene-environment interactions in the present study given the relatively small sample size. Our study differs from the SELGEN intervention in several ways which might contribute to the different findings. Besides a longer study period (26 weeks compared to 6 weeks in the SELGEN study), different sources of selenium and dosage levels were used in the interventions. We supplemented with a dietary source of selenium, while selenite was used in the SELGEN study, and at a dosage level that was half the one given by Méplan et al. [19]. Compared to inorganic selenium, organic selenium compounds, such as selenomethionine present in fish and mussels, are non-specifically incorporated into the general protein pool, leaving a smaller part of the absorbed selenium readily available for biosynthesis of selenoproteins. However, it is not clear whether this had an impact on the results, given the long supplementation period and the relatively high bioavailability of selenium from fish which has been reported to be between 56 and 90% [11, 12, 38].

Increasing the intake of selenium with fish and mussels did not affect erythrocyte GPX enzyme activity, suggesting that this selenoprotein was maximally expressed prior to supplementation. This is consistent with previous results, showing that the antioxidant enzyme GPX is saturated at blood selenium concentrations around 100 ng/mL [27, 46]. When stratifying the data according to genotype, in line with previous findings [34], carriers of the variant GPX1 allele had significantly lower erythrocyte GPX activity compared to CC homocygotes. A compromised antioxidant defense may lead to increased susceptibility to oxidative stress and DNA damage [25].

The increase in selenoprotein P concentration in the control group from week 13 to week 26 is unexpected and is difficult to explain (Fig. 2). This has been discussed in more detail in the main paper that evaluated the intervention effect on blood selenium status [28].

Strengths and limitations of the overall study design were thoroughly described by Outzen et al. [28]. Shortly, the strengths of the study are the design with the randomization procedure, which ensure evenly distribution of confounders, the restrictive inclusion and exclusion criteria, and the long-term duration of the intervention combined with the high compliance. Limitations include lack of blinding and that participants were non-fasting at blood sampling. However, we based our results on whole blood which has been shown to reflect the long-term selenium status [1, 17] as opposed to plasma selenium, which has a short half-life [5]. With regard to the lack of blinding, this is not expected to influence on this study since we mainly studied the effect of genetic variation in the intervention group. We are well aware of the present study being underpowered due to the small sample size. Nevertheless, our study is in accordance with the SELGEN study, and we are therefore able to support the notion that variation in the SELENOP gene affects selenium biomarker concentration after intake of both a selenium dietary supplement and food with a high content of selenium.

Conclusion

Taken together, our study indicates that genetically determined variation in SELENOP leads to different responses in expression of selenoproteins following consumption of selenium-rich foods. This emphasizes the importance of taking individual aspects such as genotypes into consideration when assessing risk in public health recommendations, since they may affect selenium metabolism and response to selenium intake and thereby enable a more personalized approach to micronutrient requirements.

Notes

Acknowledgements

The authors thank Lene Svensson and Annette Landin (Technical University of Denmark, National Food Institute) for technical assistance with determining GPX enzyme activity and DNA extraction and Erik Huusfeldt Larsen for performing plasma selenoprotein P, plasma selenium, and whole blood selenium measurements.

Funding

The dietary intervention study was supported by funds from the Bjarne Saxhof Foundation. The Foundation had no influence on the design and conduct of the study, management, analysis, and interpretation of the data or preparation of the manuscript.

Availability of data and materials

The datasets used for analysis during the current study are not publicly available due to existing Danish data protection law, but available from the corresponding author on reasonable request.

Authors’ contributions

TIK, MO, GRH, and UV conceived the study. TIK extracted DNA, genotyped the participants, analyzed the data, and wrote the first draft of the manuscript. MO and AO designed and conducted the randomized intervention study. GRH analyzed GPX enzyme activities and participated in writing the first draft of the manuscript. All authors have contributed to the interpretation of data, critically revised the manuscript for important intellectual content, and approved the final manuscript.

Ethics approval and consent to participate

The study was approved by the Danish Regional Committee on Biomedical Research Ethics for the Capital Region of Denmark (H-2-2010-033) in accordance with the Helsinki declaration. It was also approved by the Danish Data Protection Agency (2010-41-5111). All participants were given oral and written information about the study; written informed consent was obtained from all participants included in the study. This study is registered at http://www.clinicaltrials.gov as DCS-53227244.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary material

12263_2018_608_MOESM1_ESM.docx (37 kb)
Additional file 1: Flow chart of study participants as previously published [28]. (DOCX 37 kb)
12263_2018_608_MOESM2_ESM.docx (24 kb)
Additional file 2: Association between mean concentrations of erythrocyte GPX enzyme activity, whole blood selenium and selenoprotein P in relation to the studied polymorphisms, and within-subject effects between genotype and time in the control group. (DOCX 24 kb)

References

  1. 1.
    Ashton K, Hooper L, Harvey LJ, et al. Methods of assessment of selenium status in humans: a systematic review. Am J Clin Nutr. 2009;89:2025S–39S.  https://doi.org/10.3945/ajcn.2009.27230F.CrossRefPubMedGoogle Scholar
  2. 2.
    Bermano G, Arthur JR, Hesketh JE. Role of the 3′ untranslated region in the regulation of cytosolic glutathione peroxidase and phospholipid-hydroperoxide glutathione peroxidase gene expression by selenium supply. Biochem J. 1996;320(3):891–5.CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Blein S, Berndt S, Joshi AD, et al. Factors associated with oxidative stress and cancer risk in the Breast and Prostate Cancer Cohort Consortium. Free Radic Res. 2014;48:380–6.  https://doi.org/10.3109/10715762.2013.875168.CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Brigelius-Flohé R, Müller C, Menard J, et al. Functions of GI-GPx: lessons from selenium-dependent expression and intracellular localization. BioFactors. 2001;14:101–6.  https://doi.org/10.1002/biof.5520140114.CrossRefPubMedGoogle Scholar
  5. 5.
    Bügel S, Larsen EH, Sloth JJ, et al. Absorption, excretion, and retention of selenium from a high selenium yeast in men with a high intake of selenium. Food Nutr Res. 2008;52:1642.  https://doi.org/10.3402/fnr.v52i0.1642.CrossRefGoogle Scholar
  6. 6.
    Combs GF Jr, Watts JC, Jackson MI, et al. Determinants of selenium status in healthy adults. Nutr J. 2011;10:75.  https://doi.org/10.1186/1475-2891-10-75.CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Cooper ML, Adami H-O, Grönberg H, et al. Interaction between single nucleotide polymorphisms in selenoprotein P and mitochondrial superoxide dismutase determines prostate cancer risk. Cancer Res. 2008;68:10171–7.  https://doi.org/10.1158/0008-5472.CAN-08-1827.CrossRefPubMedPubMedCentralGoogle Scholar
  8. 8.
    Cox DG, Tamimi RM, Hunter DJ. Gene x Gene interaction between MnSOD and GPX-1 and breast cancer risk: a nested case-control study. BMC Cancer. 2006;6:217.  https://doi.org/10.1186/1471-2407-6-217.CrossRefPubMedPubMedCentralGoogle Scholar
  9. 9.
    Fairweather-Tait SJ, Bao Y, Broadley MR, et al. Selenium in human health and disease. Antioxid redox signal. 2011;14:1337–83.  https://doi.org/10.1089/ars.2010.3275.CrossRefPubMedGoogle Scholar
  10. 10.
    Forsberg L, De Faire U, Morgenstern R. Low yield of polymorphisms from EST blast searching: analysis of genes related to oxidative stress and verification of the P197L polymorphism in GPX1. Hum Mutat. 1999;13:294–300.  https://doi.org/10.1002/(SICI)1098-1004(1999)13:4<294::AID-HUMU6>3.0.CO;2-5.CrossRefPubMedGoogle Scholar
  11. 11.
    Fox TE, Atherton C, Dainty JR, et al. Absorption of selenium from wheat, garlic, and cod intrinsically labeled with Se-77 and Se-82 stable isotopes. Int J Vitam Nutr Res. 2005;75:179–86.  https://doi.org/10.1024/0300-9831.75.3.179.CrossRefPubMedGoogle Scholar
  12. 12.
    Fox TE, Van den Heuvel EGHM, Atherton CA, et al. Bioavailability of selenium from fish, yeast and selenate: a comparative study in humans using stable isotopes. Eur J Clin Nutr. 2004;58:343–9.  https://doi.org/10.1038/sj.ejcn.1601787.CrossRefPubMedGoogle Scholar
  13. 13.
    Geybels MS, van den Brandt PA, Schouten LJ, et al. Selenoprotein gene variants, toenail selenium levels, and risk for advanced prostate cancer. J Natl Cancer Inst. 2014;106:dju003.  https://doi.org/10.1093/jnci/dju003.CrossRefPubMedGoogle Scholar
  14. 14.
    Hurst R, Collings R, Harvey LJ, et al. EURRECA—estimating selenium requirements for deriving dietary reference values. Crit Rev Food Sci Nutr. 2013;53:1077–96.  https://doi.org/10.1080/10408398.2012.742861.CrossRefPubMedGoogle Scholar
  15. 15.
    J a N, Batista BL, Rodrigues JL, et al. A simple method based on Icp-Ms for estimation of background levels of arsenic, cadmium, copper, manganese, nickel, lead, and selenium in blood of the Brazilian population. J Toxicol Environ Heal Part A. 2010;73:878–87.  https://doi.org/10.1080/15287391003744807.CrossRefGoogle Scholar
  16. 16.
    Karunasinghe N, Han DY, Zhu S, et al. Serum selenium and single-nucleotide polymorphisms in genes for selenoproteins: relationship to markers of oxidative stress in men from Auckland, New Zealand. Genes Nutr. 2012;7:179–90.  https://doi.org/10.1007/s12263-011-0259-1.CrossRefPubMedGoogle Scholar
  17. 17.
    Longnecker MP, Stram DO, Taylor PR, et al. Use of selenium concentration in whole blood, serum, toenails, or urine as a surrogate measure of selenium intake. Epidemiology. 1996;7:384–90.  https://doi.org/10.1097/00001648-199607000-00008.CrossRefPubMedGoogle Scholar
  18. 18.
    Majolo F, de Oliveira Paludo FJ, Ponzoni A, et al. Effect of 593C> T GPx1 SNP alone and in synergy with 47C> T SOD2 SNP on the outcome of critically ill patients. Cytokine. 2015;71:312–7.  https://doi.org/10.1016/j.cyto.2014.10.020.CrossRefPubMedGoogle Scholar
  19. 19.
    Méplan C, Crosley LK, Nicol F, et al. Genetic polymorphisms in the human selenoprotein P gene determine the response of selenoprotein markers to selenium supplementation in a gender-specific manner (the SELGEN study). FASEB J. 2007;21:3063–74.CrossRefPubMedGoogle Scholar
  20. 20.
    Meplan C, Crosley LK, Nicol F, et al. Functional effects of a common single-nucleotide polymorphism (GPX4c718t) in the glutathione peroxidase 4 gene: interaction with sex. Am J Clin Nutr. 2008;87:1019–27.CrossRefPubMedGoogle Scholar
  21. 21.
    Méplan C, Dragsted LO, Ravn-Haren G, et al. Association between polymorphisms in glutathione peroxidase and selenoprotein P genes, glutathione peroxidase activity, HRT use and breast cancer risk. PLoS One. 2013;8:e73316.  https://doi.org/10.1371/journal.pone.0073316.CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Meplan C, Hughes DJ, Pardini B, et al. Genetic variants in selenoprotein genes increase risk of colorectal cancer. Carcinogenesis. 2010;31:1074–9.  https://doi.org/10.1093/carcin/bgq076.CrossRefPubMedGoogle Scholar
  23. 23.
    Méplan C, Nicol F, Burtle BT, et al. Relative abundance of selenoprotein P isoforms in human plasma depends on genotype, se intake, and cancer status. Antioxid Redox Signal. 2009;11:2631–40.  https://doi.org/10.1089/ars.2009.2533.CrossRefPubMedGoogle Scholar
  24. 24.
    Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 1988;16:1215.CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Miranda-Vilela AL, Alves PC, Akimoto AK, et al. Gene polymorphisms against DNA damage induced by hydrogen peroxide in leukocytes of healthy humans through comet assay: a quasi-experimental study. Environ Health. 2010;9:21.  https://doi.org/10.1186/1476-069X-9-21.CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    Nemoto M, Nishimura R, Sasaki T, et al. Genetic association of glutathione peroxidase-1 with coronary artery calcification in type 2 diabetes: a case control study with multi-slice computed tomography. Cardiovasc Diabetol. 2007;6:23.  https://doi.org/10.1186/1475-2840-6-23.CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Nève J. Methods in determination of selenium states. J Trace Elem Electrolytes Health Dis. 1991;5:1–17.PubMedGoogle Scholar
  28. 28.
    Outzen M, Tjønneland A, Larsen EH, et al. The effect on selenium concentrations of a randomized intervention with fish and mussels in a population with relatively low habitual dietary selenium intake. Nutrients. 2015a;7:608–24.  https://doi.org/10.3390/nu7010608.CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
    Outzen M, Tjønneland A, Larsen EH, et al. Effect of increased intake of fish and mussels on exposure to toxic trace elements in a healthy, middle-aged population. Food Addit Contam Part A. 2015b;32:1858–66.  https://doi.org/10.1080/19440049.2015.1072878.CrossRefGoogle Scholar
  30. 30.
    Papp LV, Lu J, Holmgren A, Khanna KK. From selenium to selenoproteins: synthesis, identity, and their role in human health. Antioxid Redox Signal. 2007;9:775–806.  https://doi.org/10.1089/ars.2007.1528.CrossRefPubMedGoogle Scholar
  31. 31.
    Raaschou-Nielsen O, Soerensen M, Hansen RDRD, et al. GPX1 Pro198Leu polymorphism, interactions with smoking and alcohol consumption, and risk for lung cancer. Cancer Lett. 2007;247:293–300.  https://doi.org/10.1016/j.canlet.2006.05.006.CrossRefPubMedGoogle Scholar
  32. 32.
    Ravn-Haren G, Bügel S, Krath BN, et al. A short-term intervention trial with selenate, selenium-enriched yeast and selenium-enriched milk: effects on oxidative defence regulation. Br J Nutr. 2008a;99(4):883–92.  https://doi.org/10.1017/S0007114507825153.CrossRefPubMedGoogle Scholar
  33. 33.
    Ravn-Haren G, Krath BN, Overvad K, et al. Effect of long-term selenium yeast intervention on activity and gene expression of antioxidant and xenobiotic metabolising enzymes in healthy elderly volunteers from the Danish Prevention of Cancer by Intervention by Selenium (PRECISE) pilot study. Br J Nutr. 2008b;99:1190–8.  https://doi.org/10.1017/S0007114507882948.PubMedGoogle Scholar
  34. 34.
    Ravn-Haren G, Olsen A, Tjonneland A, et al. Associations between GPX1 Pro198Leu polymorphism, erythrocyte GPX activity, alcohol consumption and breast cancer risk in a prospective cohort study. Carcinogenesis. 2006;27:820–5.CrossRefPubMedGoogle Scholar
  35. 35.
    Rayman MP. Food-chain selenium and human health: emphasis on intake. Br J Nutr. 2008;100(2):254–68.  https://doi.org/10.1017/S0007114508939830.PubMedGoogle Scholar
  36. 36.
    Rayman MP. Selenoproteins and human health: insights from epidemiological data. Biochim Biophys Acta - Gen Subj. 2009;1790:1533–40.CrossRefGoogle Scholar
  37. 37.
    Rayman MP. Selenium and human health. Lancet. 2012;379:1256–68.CrossRefPubMedGoogle Scholar
  38. 38.
    Robinson MF, Rea HM, Friend GM, et al. On supplementing the selenium intake of New Zealanders. 2. Prolonged metabolic experiments with daily supplements of selenomethionine, selenite and fish. Br J Nutr. 1978;39:589–600.CrossRefPubMedGoogle Scholar
  39. 39.
    Rosenberger A, Illig T, Korb K, et al. Do genetic factors protect for early onset lung cancer? A case control study before the age of 50 years. BMC Cancer. 2008;8:60.  https://doi.org/10.1186/1471-2407-8-60.CrossRefPubMedPubMedCentralGoogle Scholar
  40. 40.
    Saito Y, Sato N, Hirashima M, et al. Domain structure of bi-functional selenoprotein P. Biochem J. 2004;381:841–6.  https://doi.org/10.1042/BJ20040328.CrossRefPubMedPubMedCentralGoogle Scholar
  41. 41.
    Saxholt E, Christensen AT, Møller A, et al (2008) Danish Food Composition Databank, revision 7. In: Dep. Nutr. Natl. Food Institute, Tech. Univ. Denmark. http://www.foodcomp.dk/
  42. 42.
    Soerensen M, Christensen K, Stevnsner T, Christiansen L. The Mn-superoxide dismutase single nucleotide polymorphism rs4880 and the glutathione peroxidase 1 single nucleotide polymorphism rs1050450 are associated with aging and longevity in the oldest old. Mech Ageing Dev. 2009;130:308–14.  https://doi.org/10.1016/j.mad.2009.01.005.CrossRefPubMedPubMedCentralGoogle Scholar
  43. 43.
    Steinbrecher A, Méplan C, Hesketh J, et al. Effects of selenium status and polymorphisms in selenoprotein genes on prostate cancer risk in a prospective study of European men. Cancer Epidemiol Biomark Prev. 2010;19:2958–68.  https://doi.org/10.1158/1055-9965.EPI-10-0364.CrossRefGoogle Scholar
  44. 44.
    Strauss E, Oszkinis G, Staniszewski R. SEPP1 gene variants and abdominal aortic aneurysm: gene association in relation to metabolic risk factors and peripheral arterial disease coexistence. Sci Rep. 2014;4:7061.  https://doi.org/10.1038/srep07061.CrossRefPubMedPubMedCentralGoogle Scholar
  45. 45.
    Tang TS, Prior SL, Li KW, et al. Association between the rs1050450 glutathione peroxidase-1 (C > T) gene variant and peripheral neuropathy in two independent samples of subjects with diabetes mellitus. Nutr Metab Cardiovasc Dis. 2012;22:417–25.  https://doi.org/10.1016/j.numecd.2010.08.001.CrossRefPubMedGoogle Scholar
  46. 46.
    Thomson CD. Assessment of requirements for selenium and adequacy of selenium status: a review. Eur J Clin Nutr. 2004;58:391–402.  https://doi.org/10.1038/sj.ejcn.1601800.CrossRefPubMedGoogle Scholar
  47. 47.
    Udler M, Maia AT, Cebrian A, et al. Common germline genetic variation in antioxidant defense genes and survival after diagnosis of breast cancer. J Clin Oncol. 2007;25:3015–23.CrossRefPubMedGoogle Scholar
  48. 48.
    Villette S, Kyle JAM, Brown KM, et al. A novel single nucleotide polymorphism in the 3′ untranslated region of human glutathione peroxidase 4 influences lipoxygenase metabolism. Blood Cells Mol Dis. 2002;29:174–8.  https://doi.org/10.1006/bcmd.2002.0556.CrossRefPubMedGoogle Scholar
  49. 49.
    Wheeler CR, Salzman JA, Elsayed NM, et al. Automated assays for superoxide dismutase, catalase, glutathione peroxidase, and glutathione reductase activity. Anal Biochem. 1990;184:193–9.CrossRefPubMedGoogle Scholar
  50. 50.
    Ye H, Li X, Wang L, et al. Genetic associations with coronary heart disease: meta-analyses of 12 candidate genetic variants. Gene. 2013;531:71–7.  https://doi.org/10.1016/j.gene.2013.07.029.CrossRefPubMedGoogle Scholar

Copyright information

© The Author(s) 2018

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors and Affiliations

  1. 1.National Food InstituteTechnical University of DenmarkKgs LyngbyDenmark
  2. 2.Danish Cancer Society Research CenterCopenhagen ØDenmark
  3. 3.The Danish Multiple Sclerosis RegistryCopenhagen University Hospital, RigshospitaletCopenhagen ØDenmark
  4. 4.National Research Centre for the Working EnvironmentCopenhagen ØDenmark
  5. 5.The Danish Multiple Sclerosis Center, Department of Neurology, The Danish Multiple Sclerosis Registry, Section 7801RigshospitaletCopenhagen ØDenmark

Personalised recommendations