Mol. Nutr. Food Res.

2013, 57, 21952205

2195

DOI 10.1002/mnfr.201300168

RESEARCH ARTICLE

Selenium alters miRNA profile in an intestinal cell line:

Evidence that miR-185 regulates expression of GPX2
and SEPSH2
Anabel Maciel-Dominguez1,2 , Daniel Swan1 , Dianne Ford1,2 and John Hesketh1,2
1
2

Institute for Cell and Molecular Biosciences, Newcastle University, Newcastle upon Tyne, UK
Human Nutrition Research Centre, Newcastle University, Newcastle upon Tyne, UK

Scope: Intake of the essential micronutrient selenium (Se) has health implications. This work
addressed whether some effects of Se on gene expression are exerted through microRNAs
(miRNA).
Methods and results: Human colon adenocarcinoma cells (Caco-2) were grown in Se-deficient
or Se-adequate medium for 72 h. RNA was extracted and subjected to analysis of 737 miRNA
using microarray technology. One hundred and forty-five miRNA were found to be expressed in
Caco-2 cells. Twelve miRNA showed altered expression after Se depletion: miR-625, miR-492,
miR-373*, miR-22, miR-5325p, miR-106b, miR-30b, miR-185, miR-203, miR1308, miR-285p,
miR-10b. These changes were validated by quantitative real-time PCR (RT-qPCR). Transcriptomic analysis showed that Se depletion altered expression of 50 genes including selenoproteins
GPX1, SELW, GPX3, SEPN1, SELK, SEPSH2 and GPX4. Pathway analysis identified arachidonic acid metabolism, glutathione metabolism, oxidative stress, positive acute phase response
proteins and respiration of mitochondria as Se-sensitive pathways. Bioinformatic analysis identified 13 transcripts as targets for the Se-sensitive miRNA; three were predicted to be recognised
by miR-185. Silencing of miR-185 increased GPX2 and SEPSH2 expression.
Conclusions: We propose that miR-185 plays a role in up-regulation of GPX2 and SEPHS2
expression. In the case of SEPHS2 this may contribute to maintaining selenoprotein synthesis.
The data indicate that micronutrient supply can regulate the cell miRNA expression profile.

Received: March 7, 2013

Revised: April 25, 2013
Accepted: May 23, 2013

Keywords:
Epithelial cells / Micronutrient / MicroRNA / Selenoprotein / Transcriptomics


1

Additional supporting information may be found in the online version of this article at
the publishers web-site

Introduction

Sub-optimal micronutrient intake has been proposed to affect health during ageing [1]. Selenium (Se), zinc and vitaCorrespondence: Dr. John Hesketh, Institute for Cell and Molecular Biosciences, Newcastle University, Newcastle-upon-Tyne, NE2
4HH, UK
E-mail: j.e.hesketh@ncl.ac.uk
Fax: +44-1912227424
Abbreviations: Caco-2, human colon adenocarcinoma cells;
Ct, cycle threshold value; FDR, false discovery rate; GAPDH,
glyceraldehyde 3-phosphate dehydrogenase; IPA, ingenuity
pathway analysis; miRNA, microRNA; MRE, microRNA recognition element; NMD, nonsense-mediated decay; RISC, RNAinduced silencing complex; RT-qPCR, quantitative real-time
PCR; Se, selenium; Sec-tRNA, selenocysteine tRNA

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

min D have important metabolic roles that are reflected in

their effects on cell growth and proliferation and in their
sub-optimal intake impacting on human health [24]. Transcriptomic analysis has shown that for each of these micronutrients changes in intake lead to altered gene expression in
several biochemical pathways and networks [58]. However,
despite these known effects and the proposed role of microRNAs (miRNA) as regulators of gene networks [9], there is little
information available on whether epigenetic mechanisms involving regulation through miRNA play a role in metabolic
regulation by micronutrients. However, recently zinc depletion was reported to alter blood miRNA levels [10] and miR-22
has been proposed to be a target of 1,25-dihydroxyvitamin
D3 in colon cancer cells [11]. The present work investigated
whether changes in Se supply affected miRNA expression
in human colon adenocarcinoma cells (Caco-2), an intestinal
www.mnf-journal.com

2196

A. Maciel-Dominguez et al.

cell line. The focus on Se and intestinal cells arose from observations that Se intake affects colorectal adenoma and cancer
risk (see [1214]) and the reported importance of Se and Secontaining selenoproteins in inflammatory mechanisms in
the gut [1517].
Se is incorporated into selenoproteins as the amino-acid
selenocysteine using a specific tRNA (Sec-tRNA), a structure
within the 3 untranslated region (3 UTR) of the mRNA (the
SECIS) and a UGA codon [18]. The selenoproteins include
the glutathione peroxidases (GPx1, GPx2 and GPx4), which
protect cells from reactive oxygen species, the thioredoxin
reductases, which function in redox control, selenoproteins
N and S, which have roles in the endoplasmic reticulum
and selenoprotein P, which is a Se-transport protein [18, 19].
Changes in Se intake alter the levels and activity of selenoproteins, and critically the effects differ between different
selenoproteins and different tissues. This results in a hierarchy of selenoproteins in which some are more sensitive to
changes in intake than others [18, 19]. The mechanistic basis
for this hierarchy is unclear but both differences in 3 UTR
sequences and methylation of Sec-tRNA are thought to be
involved [2022]. Se intake also affects the activity of other
biochemical pathways, presumably secondary effects due to
metabolic changes as a result of altered selenoprotein expression. For example, transcriptomic studies show that moderate Se deficiency leads to alterations in inflammatory, endoplasmic reticulum unfolded protein response and protein
biosynthetic pathways in the mouse colon [5]. In humans Se
supplementation changes protein biosynthetic pathways [23].
However, the mechanisms by which Se intake influences the
expression of these other genes are largely unknown.
miRNA are small, 2030 nt long, non protein-coding
RNAs that can regulate gene expression through the formation of an RNA-induced silencing complex (RISC) [24], which
controls genome function through mRNA degradation or
translational repression [25]. miRNA recognise complementary miRNA recognition elements (MRE) throughout mRNA
sequences, including the 3 and 5 UTRs. miRNA are flexible
in their recognition of target sequences and this allows them
to regulate groups of genes [9, 2426]. One mRNA may contain multiple MRE that recognise different miRNA resulting
in the potential for expression to be regulated by multiple
miRNA. In addition, several mRNAs may contain MRE for
one miRNA resulting in the potential for one miRNA to regulate several genes within a biochemical pathway or network.
For example, miR-19a has been suggested to regulate several
genes with roles in apoptosis [26] and the mir-1792 family,
which have MRE for different protein coding genes [27], represent miRNA that target a group of genes. Such observations
are consistent with an overall model of regulation at the level
of pathways.
No information is available in the literature as to whether
changes in Se intake and the subsequent metabolic changes
lead to alterations in miRNA expression or changes in target
gene expression through miRNA. Two observations led us to
hypothesise that changes in Se supply would lead to miRNA

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Mol. Nutr. Food Res. 2013, 57, 21952205

regulation of gene expression. First, transcriptomic analyses

have shown that Se deficiency or supplementation leads to
changes in expression of multiple genes within a pathway,
e.g. the protein synthesis pathway/translation factors [5, 23].
Second, protein binding to the 3 UTR, the region of mRNAs
most frequently reported to be a target for miRNA recognition, is central to selenoprotein synthesis [18, 19]. Therefore
the aim of the present work was to use a transcriptomic approach for the analysis of the response of both the miRNA
profile and mRNA profile of an intestinal cell line to changes
in Se supply.

Materials and methods

2.1 Cell culture and RNA analysis

Caco-2 cells were grown at 37C in the presence of 5% CO2
in DMEM with 4.5 g/L glucose and Glutamax, 10% fetalcalf serum and 1% penicillin/streptomycin (100 units/mL
and 100 g/mL, respectively). Se-adequate or Se-deficient
medium comprised DMEM (with 4.5 g/L glucose and Glutamax) without added fetal-calf serum but containing 5 g/mL
insulin, 5 g/mL transferrin, 1% penicillin/streptomycin
and with or without 7 ng/mL sodium selenite (40.5 nM).
Cells were grown in Se-deficient or Se-adequate medium for
72 h. Under all these culture conditions the Caco-2 cells were
undifferentiated. RNA was extracted by standard Trizol procedure followed by passage through a glass fibre membrane
spin column (PureLink RNA Mini Kit, Invitrogen, UK). RNA
concentration and integrity was assessed by absorbance at
R
ND-1000 and by CE on an
260/280 nm on a NanoDrop
Agilent 2100 Bioanalyzer.

2.2 miRNA microarray design and analysis

A customised human genome V2 Agilent miRNA microarray
was designed using Agilents eArray Design Tool. The array
included 737 distinct Homo sapiens miRNA and 10 negative
control miRNA. The miRNA accession numbers are stated in
Supporting Information Table 1. The array targets included:
194 miRNA identified as present in colon tissue or colon
cell lines [2632]; 292 miRNA identified by microrna.org, MicroCosm, GeneSet2MiRNA, miRWalk and Ingenuity pathway analysis (IPA) and bioinformatically predicted to target
the genes within the protein biosynthetic pathways found to
be Se-sensitive in a human supplementation trial [23]; 414
miRNA taken from the Illumina human v2 chip platform
miRNA array and PCR arrays for which mature sequences
were best characterised and some experimentally validated;
lastly, 10 miRNA from Mus musculus as an internal negative
control. The miRNA expression profiling of the customised
8 15 K slide was assessed by microarray technology using
the human genome V2 Agilent platform. Hybridisation of
100 ng of total RNA was carried out as a service by Warwick
www.mnf-journal.com

2197

Mol. Nutr. Food Res. 2013, 57, 21952205

University following the miRNA Microarray System with

miRNA Complete Labelling and Hyb Kit protocol. Spike-in
solutions were added at the labelling and hybridisation stage
with the labelled RNA put through a further purification step.
Array scan was performed using the Agilent G2562CA scanner. Differential miRNA expression analysis was performed
using t-tests in GeneSpring GX 11 (Agilent) and Benjamini
Hochberg false discovery rate (FDR) multiple testing correction where probes had a corrected p-value of <0.05.

2.3 Transcriptomic analysis

Total RNA (200 ng) was labelled and hybridised to Illumina HumanRef-8 V3 Expression BeadChips by Service XS
B.V. Scanning of the arrays was performed using their standard protocols. Raw signals were normalised with no background adjustment using Illumina Genomestudio v.2010.1
software. Data were loaded into GeneSpring GX 11 for analysis and analysed using the Illumina.SingleColor.HumanRef8_V3_0_R1_11282963_A technology annotation package for
visualisation. Differential gene expression analysis was
performed using t-tests in GeneSpring (with Benjamini
Hochberg FDR multiple testing correction) where probes
had a corrected p-value of <0.05 and supplemented with
RankProducts analysis using the RankProd package in BioConductor (probes with a p < 0.05). The list of differentially
expressed genes was analysed by the IPA tool and significant
canonical pathways identified. Genes from the data set that
met differential expression cut-offs of a 1.2-fold change in
expression with an FDR of <0.05 and were associated with a
pathway entity in the Ingenuity Knowledge Base were considered for analysis. The significance of the association between
the data set and the canonical pathway was assessed by (1)
the ratio of the number of molecules from the data set that
map to the pathway divided by the total number of molecules
that map to the canonical pathway, and (2) Fishers exact test
to calculate a p-value determining the probability that the association between the genes in the data set and the canonical
pathway is explained by chance alone.

2.4 Quantitative real-time PCR (RT-qPCR)

R
miRNA assays on an
miRNA were assayed by TaqMan
Applied Biosystems HT7900 system. miRNA were reverse
transcribed with 0.15 L 100 mM dNTPs (with dTTP), 1 L
MultiScribeTM Reverse Transcriptase, 50 U/L, 1.5 L 10 X
Reverse Transcription Buffer, 0.19 L of 20 U/L RNase Inhibitor, and 4.16 L of nuclease-free water in a total volume
of 7 L. After mixing, the reaction was placed on ice while
3 L of 5 X RT primers and 5 L of 10 ng RNA were added.
After 5 min incubation on ice, the final reaction was transferred to a thermal cycler and reverse transcription carried
out at 16C for 30 min followed by 42C for 30 min and 85C
for 5 min before being placed on ice. To control for possible


C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

variation in reverse transcription between each miRNA, the

same sample was reverse transcribed with all the different
miRNA primers sets of interest (see Supporting Information
Table 2) at the same time. Real-time PCR was carried out by
adding 1 L of specific miRNA primers corresponding to the
miRNA used in the first step of cDNA synthesis, 7.76 L of
0.05% diethylpyrocarbonate-treated, 1.33 L of RT product
R
Universal
(from cDNA synthesis), 10 L of 2 X TaqMan
R
UNG to a final volume of
PCR Master Mix No AmpErase
20 L per well. Cycle threshold values (Ct) were extracted
using a 2.0 threshold and a baseline set 23 cycles before
the amplification curve was used. Relative quantification was
calculated using the 2 Ct formula using the Ct values of the
miRNA of interest minus those of the reference miRNA.
RT-qPCR analysis of mRNA was carried out on an Applied
Biosystems HT7900 system and relative quantification was
obtained by comparing the concentration of a target gene
with that of the reference gene glyceraldehyde 3-phosphate
dehydrogenase (GAPDH) in the same sample. Melting curve
analysis of products showed a single peak across all samples
and standard curves; standard curves showed the efficiency
was close to 2.0 for all primer pairs, and error (<0.1) and slope
passed Light Cycler 480 Roche quality control (Operators
manual for Software Version 1.5). Extraction of Ct values
and their use to calculate relative levels of mRNA expression
was as described above. Primers are shown in Supporting
Information Table 3.

2.5 miRNA silencing

miRNA silencing was performed by treatment with a perfectly
complementary oligonucleotide (anti-miR) specific to miR185 (sequence: AGGGGCUGGCUUUCCUCUGGUC) or a
negative control (catalogue no. AM17010; both from Applied
Biosystems, UK); 3.5 105 cells were seeded in each well of
R
six-well plates and for each plate 5 L of Lipofectamine
2000 Transfection Reagent (Invitrogen) was added to
R
(Invitrogen) and incubated for 5 min.
250 L of Opti-MEM
At the same time either anti-MiR-185 or the negative control
R
and incubated for 5 min.
was added to 250 L of Opti-MEM
After incubation, these two mixtures were combined and incubated for 20 min and then added to the cells to give a final
concentration of 70 nM. Culture medium was changed 24 h
after transfection.

2.6 Bioinformatics
GeneSet2MiRNA (which collects information from 11 bioinformatic tools), microrna.org, MicroCosm, miRWalk (which
are based on complementary and conserved binding predictions) and IPA were used to predict both miRNA likely to be
expressed in Caco-2 cells and/or to be regulated by Se and to
predict miRNA that interact with target genes identified on
the microarray [28, 29].
www.mnf-journal.com

2198

A. Maciel-Dominguez et al.

Results

Caco-2 cells were grown in either Se-deficient medium or

Se-adequate medium containing 40 nM sodium selenite for
72 h [30]. GPX1 and SELW mRNA levels, which are known
to be responsive to alterations in Se supply [5, 7], were lower
in cells grown under Se-deficient conditions compared with
those grown in Se-adequate medium (see section 3.2) indicating that the Caco-2 cells grown in these media differ in Se
status, as observed previously [30].

3.1 miRNA expression

To determine the effect of changes in Se supply on miRNA
expression, RNA was extracted from Caco-2 cells grown in
either Se-deficient or Se-adequate medium for 72 h and hybridised to a customised human genome V2 Agilent miRNA
microarray. Data analysis using GeneSpring GX 11 indicated
that 145 miRNA were expressed in Caco-2 cells (see Supporting Information Table 4). Differential expression analysis indicated that 12 miRNA were differentially downregulated under conditions of Se depletion (Fig. 1). The greatest changes
in expression were shown by miR-185, miR-625, miR-203 and
miR-429, for which no signal was detected in RNA from Sedepleted cells. The difference in expression for the remainder
of Se-sensitive miRNA represented fold changes of 1.2.
Validation of the microarray data was carried out using quantitative realtime microRNA PCR (q-miRNA-PCR) to
measure levels of 6 of the 12 miRNA found to change on the
microarray: miR-185, miR-28-5p, miR-10b, miR-106b, miR625 and miR-429. The data were normalised using hsa-let-7a

Mol. Nutr. Food Res. 2013, 57, 21952205

as the reference miRNA since it has previously been used as a

reference in colon tissue [31,32]. After normalisation, the relative expression of the six miRNA decreased in cells grown
in Se-deficient medium compared with cells grown in Seadequate medium (Fig. 2), an observation that is consistent
with the microarray data.

3.2 Transcriptomic analysis

Aliquots of the RNA samples previously analysed by miRNA
microarray were also analysed for global gene expression using Illumina HumanRef-8 V3 Expression BeadChips. Differential gene expression analysis using a fold-change cut-off of
1.2 showed 50 genes to be differentially expressed entities (Table 1). Selenoproteins GPX1, SELW, GPX3, SEPN1, SELK and
GPX4 were downregulated at the mRNA level in Se-deficient
cells. The largest changes were in expression of GPX1 and
SELW, consistent with these being relatively low in the hierarchy and more susceptible to changes in Se supply [19,30,33].
These changes indicate that Se-depletion decreased Se status
to a degree similar to that previously observed [30]. In such
Se-depleted cells levels of GPX2 and SEPSH2 transcripts were
increased compared with Se-adequate cells.
Pathway analysis of the 50 Se-sensitive genes identified
arachidonic acid metabolism, glutathione metabolism and selenoamino acid metabolism as significantly affected canonical pathways (Table 2). Toxicity list analysis identified oxidative stress, positive acute phase response proteins and
respiration of mitochondria as being Se-sensitive processes
(Table 2). The transcriptomic microarray data were validated by analysing expression of five selected target genes by

Figure 1. miRNA expressed differentially in human intestinal Caco-2 cells in response to Se depletion. Data were generated by hybridisation
of RNA samples isolated from Caco-2 cells grown for 72 h under either Se-depleted conditions or with the addition of 7 ng/mL Se as sodium
selenite. Hybridisation was to a custom V2 Agilent 8 15 k miRNA microarray. miRNA expressed differentially were identified using a
flag filter analysis (GeneSpring GX). Data are shown as means SEM (n = 4 in each group). Levels of expression were compared by a
MannWhitney U-test, **p < 0.01, *** p < 0.001; ND = not detectable.

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mnf-journal.com

2199

Mol. Nutr. Food Res. 2013, 57, 21952205

Figure 2. Relative expression of miRNA

measured by RT-qPCR. The four RNA samples from Se-adequate and Se-depleted
Caco-2 cells previously subjected to miR
microarray analysis and two additional
RNA samples from each group were reverse transcribed using a specific stem
loop primer and amplified by hydrolysis probes. cDNA were then analysed
by qPCR. Absolute expression was normalised to let-7a and relative expression
was expressed as a percentage of the Sedepleted cells. Values shown are means
SEM (n = 6 in each group). Groups were
compared by a MannWhitney U-test,
*p < 0.05, **p < 0.01.

RT-qPCR. After normalisation of the data to GAPDH as the

reference gene, expression of GPX1, GPX3 and SELK were
found to be decreased in Se-deficient cells compared with
cells grown in Se-adequate medium, whereas expression of
SEPHS2 and PIR was increased (Fig. 3). Analysis of GPX2
was not included in this validation. These changes were comparable to those detected in the transcriptomic analysis although the changes in expression of SELK and PIR after
Se-depletion were not statistically significant.

3.3 Bioinformatic analysis of microarray results

In order to identify Se-sensitive miRNAmRNA interactions
miRNA target tools were used to compare the sequences of
the Se-sensitive gene list and the miRNA identified as responding to Se. Because miRNAmRNA cleavage occurs not
only within the 3 or 5 UTR but also in the ORF, the search for
MRE in the Se-sensitive genes was carried out without restriction to specific regions of the nucleotide sequences. Using
miRWalk, microRNA.org MicroCosm or GeneSet2miRNA 8
of the 12 Se-sensitive miRNA (miR-185, miR-625, miR429,
miR373*, miR22, miR-5325p, miR106b and miR-30b) identified by miRNAmicroarray analysis were predicted to have
target sequences in mRNAs showing differential expression
in Caco-2 cells in response to altered Se supply (Table 3).
ME1, CACYBP, ATP6V1B1, CMIP and GPX1 were predicted
to contain an MRE for miR-625. SELK was predicted to contain an MRE for miR-492. HTATIP2 and GPT2 were predicted to contain an MRE for miR-373* and miR-22. EPHX2
was predicted to contain an MRE for miR-5325p. GPX2 and
MT1F were predicted to contain an MRE for miR-106b. GPX3
was predicted to contain an MRE for miR-30b. Finally GPX2,

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

GPX3 and SEPHS2 were predicted to contain an MRE for

miR-185.

3.4 Knock-down of miR-185

Based on both the expression of miR-185 in colorectal tissue reported previously [38, 39] and the prediction that it has
MRE targets in the three Se-sensitive selenoprotein mRNAs
GPX2, GPX3 and SEPHS2, the functional link between miR185 and its predicted targets was further studied in miRNA silencing experiments. Since miR-185 expression was very low
(not detectable by microarray) in Se-deficient cells but significantly higher in Se-adequate conditions (see Figs. 1 and 2),
we predicted that it would not be possible to investigate miR185 silencing in Se-deficient cells. Therefore, miRNA knockdown experiments were carried out under Se-adequate conditions to investigate if lowering of miR-185 levels led to effects
on its predicted targets similar to those observed during Se
depletion.
Cells were grown in Se-adequate medium for 48 h with either a specific anti-miR185 or a negative control. The extent of
miRNA silencing was assessed by q-miRNA-PCR. Treatment
with 70 nM anti-miR for 48 h led to approximately 50% knockdown of miR-185 (see Supporting Information Fig. 1). Under
these conditions of miR-185 knock-down the expression of
the potential miR-185 targets GPX2 and SEPHS2 was significantly increased by approximately 60% (Fig. 4) compared
with the negative control. In contrast there was no statistically significant effect on GPX3 expression. Thus GPX2 and
SEPHS2 mRNA levels were affected by silencing of miR-185
in the direction predicted by the observed decrease in this
miRNA under the conditions of Se depletion, which caused
an increase in these transcript levels (see Table 1 and Fig. 3).
www.mnf-journal.com

2200

A. Maciel-Dominguez et al.

Mol. Nutr. Food Res. 2013, 57, 21952205

Table 1. List of genes showing altered expression in Caco-2 cells after change in Se concentration of the cell culture medium

Symbol

Entrez gene name

Accession

Fold
change

Regulation
(Se versus
Se+)

GPX1
SELW1
MT1E
MT1G
MT1F
GPX3
SELN1
MT1X
MT1A
MT1H
LOC728643
SEPHS2
CMIP
GPX2
PIR
MT2A
FABP7
CGB5
DCUN1D4

Glutathione peroxidase 1
Selenoprotein W, 1
Metallothionein 1E
Metallothionein 1G
Metallothionein 1F
Glutathione peroxidase 3
Selenoprotein N, 1
Metallothionein 1X
Metallothionein 1A
Metallothionein 1H

NM_201397.1
NM_003009.2
NM_175617.3
NM_005950.1
NM_005949.2
NM_002084.3
NM_206926.1
NM_005952.2
NM_005946.2
NM_005951.2
NR_003277.1
NM_012248.2
NM_030629.1
NM_002083.2
NM_001018109.1
NM_005953.2
NM_001446.3
NM_033043.1
NM_015115.1

3.321
2.651
1.823
1.804
1.777
1.716
1.58
1.527
1.522
1.509
1.428
1.392
1.378
1.375
1.37
1.357
1.324
1.323
1.306

NM_001979.4
NM_002395.3
NM_001018109.1
NM_001008395.2
NM_001040034.1
NM_006410.3
NM_022872.2
NM_000607.1
NM_001017998.2
NM_001730.3
NM_004751.1
NM_004821.1
NM_002483.3
NM_052831.2

1.305
1.301
1.298
1.294
1.292
1.286
1.284
1.283
1.282
1.28
1.276
1.272
1.27
1.27

NM_021237.3
NM_001039847.1
NM_003878.1

1.269
1.255
1.25

NM_080491.1
NM_002291.1
NM_001692.3
NM_000509.4
NM_007280.1
NM_021034.2
NM_014412.2
NM_177530.1

1.245
1.245
1.242
1.239
1.235
1.227
1.225
1.22

NM_021227.2
NM_004172.3

1.217
1.217

NM_006435.2
NM_002346.1
XM_001133534.1
NM_133443.1
NM_007003.2
NM_018530.2
NM_013283.3

1.211
1.208
1.205
1.204
1.197
1.179
1.135

EPHX2
ME1
PIR
C7ORF59
CD63
HTATIP2
IFI6
ORM1
GNG10
KLF5
GCNT3
HAND1
C6ORF192
CEACAM6
SELK
GPX4
GGH
GAB2
LAMB1
ATP6V1B1
FGG
OIP5
IFITM3
CACYBP
SULT1A1
OSTC
SLC1A3
IFITM2
LY6E
ATP1B3
GPT2
PAGE4
GSDMB
MAT2B

Selenophosphate synthetase 2
c-Maf-inducing protein
Glutathione peroxidase 2 (gastrointestinal)
Pirin (iron-binding nuclear protein)
Metallothionein 2A
Fatty acid binding protein 7, brain
Chorionic gonadotropin, beta polypeptide 5
DCN1, defective in cullin neddylation 1, domain containing 4
(Saccharomyces cerevisiae)
Epoxide hydrolase 2, cytoplasmic
Malic enzyme 1, NADP(+)-dependent, cytosolic
Pirin (iron-binding nuclear protein)
Chromosome 7 ORF 59
CD63 molecule
HIV-1 Tat interactive protein 2, 30 kDa
Interferon, alpha-inducible protein 6
Orosomucoid 1
Guanine nucleotide binding protein (G protein), gamma 10
Kruppel-like factor 5 (intestinal)
Glucosaminyl (N-acetyl) transferase 3, mucin type
Heart and neural crest derivatives expressed 1
Chromosome 6 ORF 192
Carcinoembryonic antigen-related cell adhesion molecule 6
(non-specific cross reacting antigen)
Selenoprotein K
Glutathione peroxidase 4 (phospholipid hydroperoxidase)
Gamma-glutamyl hydrolase (conjugase,
folylpolygammaglutamyl hydrolase)
GRB2-associated binding protein 2
Laminin, beta 1
ATPase, H+ transporting, lysosomal 56/58kDa, V1 subunit B1
Fibrinogen gamma chain
Opa interacting protein 5
Interferon induced transmembrane protein 3 (18U)
Calcyclin binding protein
Sulfotransferase family, cytosolic, 1A, phenol-preferring,
member 1
Oligosaccharyltransferase complex subunit
Solute carrier family 1 (glial high affinity glutamate transporter),
member 3
Interferon induced transmembrane protein 2 (18D)
Lymphocyte antigen 6 complex, locus E
Glutamic pyruvate transaminase (alanine aminotransferase) 2
P antigen family, member 4 (prostate associated)
Gasdermin B
Methionine adenosyltransferase II,


C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mnf-journal.com

2201

Mol. Nutr. Food Res. 2013, 57, 21952205

Table 2. Pathway analysis of transcriptomic data from Caco-2

cells grown under different Se concentrations in the cell
culture medium

Canonical pathway

p-Value

Ratio of
Se-sensitive
genes to
total
number
of genes

Table 3. Bioinformatic analysis to identify targets of miRNA

sensitive to cell culture Se concentrations

miRNA

miRNA target

Bioinformatic tool

hsa-miR-185

GPX3

miRWalk
microRNA.org
MicroCosm
microRNA.org
microRNA.org
miRWalk
GeneSet2miRNA
mirWalk

hsa-miR-625
Arachidonic acid metabolism
Glutathione metabolism
Selenoamino acid metabolism
Extrinsic prothrombin
activation pathway
Folate biosynthesis
Sulfur metabolism
Toxicity mechanisms
Oxidative stress
Positive acute phase
response proteins
Decreased respiration of
mitochondria

2.8E-05
3.6E-05
6.5E-03
5.3E-02

5/113 (0.044)
4/59 (0.068)
2/38 (0.053)
1/17 (0.59)
hsa-miR-429

5.3 E-02
4.4E-02

1/17 (0.059)
1/10 (0.1)

8.7E-07
4.6E-03

5/56 (0.089)
2/32 (0.062)

3.4E-02

1/11 (0.091)

The lack of effect of miR-185 silencing on GPX3 transcript

levels was consistent with the opposite (decrease) in GPX3
expression observed after Se depletion (Fig. 3).

Discussion

The present results show for the first time that changes in
Se supply not only lead to changes in expression of protein
coding genes [5, 7, 23] but also alter the miRNA profile in a
mammalian cell line, in particular a gastrointestinal cell line.
Furthermore the changes in the miRNA profile are paralleled
by changes in expression of a number of target genes predicted to contain MRE for those miRNA. The response of two
predicted mRNA targets for a specific miRNA (miR-185) se-

hsa-miR-373*
hsa-miR-22
hsa-miR-5325p
hsa-miR-106b
hsa-miR-30b

GPX2
SEPHS2
ME1
CACYBP
ATP6V1B1
CMIP
GPX1
SELK
HTATIP2
GPT2
GTP2
EPHX2
GPX2
MT1F
GPX3

MicroCosm
miRWalk
microRNA.org
miRWalk
miRWalk
miRWalk
MicroCosm
microRNA.org
MicroCosm

lected to test the relationship were consistent with this mode

of regulation.
From the 737 entities included on the miRNA chip, 145
miRNA were expressed in Caco-2 cells. Seventy-one of these
were present in the list of 194 miRNA that had previously
been found to be expressed in intestinal cells or tissues. Fifteen miRNA expressed were predicted bioinformatically and
do not correspond to any miRNA previously described in
the intestinal microRNAome or included on PCR arrays or
microarrays. Expression of 12 miRNA was found to change
following Se-depletion. Little is known about the Se-sensitive
miRNA identified, especially as regards their function in intestinal cells. miR-429 is expressed at low levels in colon cell
lines [34, 35] and has been suggested to be a colorectal cancer biomarker [35]. High levels of miR-10b expression have

Figure 3. Relative expression of target

mRNAs using RT-qPCR. RNA samples
from Caco-2 cells grown in Se-adequate
or Se-depleted medium (for 72 h) corresponded to those used for miR analysis
(n = 6 in each group). RNA was reverse
transcribed and then analysed by qPCR
for levels of five targets found to be differentially expressed by microarray analysis (GPX1, GPX3, SEPHS2, SELK and PIR).
Data were normalised to GAPDH. GPX1,
GPX3 and SELK were found to be upregulated in cells grown in Se-adequate
medium, whereas SEPHS2 and PIR were
downregulated in cells cultured under the
same conditions. Values shown are means
SEM. Groups were compared by a
MannWhitney U test, ***p < 0.001.

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mnf-journal.com

2202

A. Maciel-Dominguez et al.

Mol. Nutr. Food Res. 2013, 57, 21952205

Figure 4. Effects of miR-185 silencing on

target gene expression in Caco-2 cells.
mRNA levels of SEPHS2, GPX2 and GPX3
were measured by RT-qPCR in RNA from
cells grown in Se-adequate medium after
48 h of anti-miR exposure. Expression was
normalised to GAPDH. The expression of
GPX2 and SEPHS2 changed significantly
at 48 h of anti-miR185 exposure; the effects on GPX3 were not significant. Values
shown are means SEM (n = 9). Groups
were compared by a MannWhitney Utest, **p < 0.01, **p < 0.001.

been related to malignancy in colorectal cancer samples [36]

and miR-196b has been detected in gastric and colon cancer
biopsies [37, 38], as well as in circulating blood from cancer patients [39]. miR-203 has also been detected in colorectal
biopsies, and studies in HT29 cells suggest it has tumour suppressor properties [39]. In HT29 cells, miR-22 over-expression
has been linked to apoptosis [40]. It has recently been suggested that miR-185 has a role in colorectal cancer progression, with metastasis and low cancer survival being associated
with over-expression [41]. Of the 12 miRNA found to change
expression in the present study, miR-185 showed one of the
biggest changes in expression after Se-depletion. However,
the expression of miR-7, previously reported to target the
5 UTR of selenoprotein P variant 1b transcripts [42], was not
changed in Se-depleted Caco-2 cells.
miR-185 was predicted to target three genes that code for
selenoproteins. Of these three, the target SEPHS2 was shown
by both microarray and real-time PCR analysis to express increased mRNA levels after Se-depletion. In addition, the microarray analysis showed the target GPX2 to also be expressed
at increased levels, as has been observed previously in Caco-2
cells [30]. Under the same conditions miR-185 levels were
decreased and therefore we predicted that the abundance of
these mRNAs would be increased by knock-down of mi-185.
In contrast, for GPX3 the response to Se depletion was in
the same (negative) direction as the response of the miRNA
and thus the prediction was that GPX3 mRNA abundance
would be unaffected by knock-down of mi-185. Effects mediated through miRNA would not be expected to account for
all responses of selenoprotein genes to Se levels, and we propose that the observed response of GPX3 is one such example.
However, we cannot exclude the possibility that lack of effect
on GPX3 was due to subtle differences in the sequences of the
three target MRE. As predicted, knockdown of mi-185 had no

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

effect on GPX3 transcript levels but did increase levels of both

GPX2 and SEPHS2. These observations are consistent with a
feedback control mechanism in which miR-185 has a negative
regulatory effect on GPX2 and SEPHS2 expression. We hypothesise that Se supply affects selenoprotein activities which
in turn causes unknown biochemical changes that affect miR185 expression and this regulates GPX2 and SEPHS2. Both
GPX2 and SEPSH2 are high in the selenoprotein hierarchy
and their expression is little affected, or increased, by lower
Se supply [30,43,44]. We propose that regulation of these two
selenoprotein genes by miR-185 contributes to their different response to lower Se intake compared with that of other
selenoproteins. miR-185 has been previously reported to be
expressed in colorectal tissue [38, 39], and thus it is possible that in the colon miR185 is involved in feedback control
of SEPSH2 and GPX2 expression. Interestingly, in prostate
cells miR-17* has been reported to suppress mitochondrial
antioxidant enzymes and GPX2 in prostatic cells [45] but this
miRNA was not expressed in Caco-2.
Interestingly, in SEPSH2 the putative MRE for miR-185
lies within the sequence predicted to form the SECIS element. Since the SECIS functions by recruiting binding proteins into the Secys-tRNA incorporation complex [18, 44], it
is possible that miR-185 could regulate SEPHS2 expression
not only by regulating mRNA degradation through the RISC
complex but also through modulation of selenocysteine incorporation. The effect of miR-185 knockdown on GPX2 transcript levels suggests that its effects are on mRNA degradation. Since it has been suggested that nonsense-mediated
decay (NMD) may play a role in selenoprotein mRNA degradation [46], it is possible that in the present work NMD
may have influenced the effects of Se-depletion on selenoprotein transcript levels. In addition, it is not known to
what extent miRNA regulation of degradation pathways and
www.mnf-journal.com

2203

Mol. Nutr. Food Res. 2013, 57, 21952205

NMD are inter-related. There is debate over the importance

of NMD in selenoprotein regulation [18, 47] and so future
work is needed to assess the relative importance of miRNA
and NMD in the changes in the hierarchy of selenoprotein
expression.
Se-depletion of Caco-2 cells changed expression of specific
selenoprotein mRNAs, consistent with previous data showing Se depletion to affect expression of GPX1, SELW, GPX4
and SEPSH2 in several cell lines and tissues including Caco-2
cells [30]. Interestingly, of these, GPX1 and GPX4 have been
reported to be up-regulated when cancer and healthy colonic
tissue were compared [48]. The sensitivity of SELK expression to Se supply (Table 1) is consistent with data from the
mouse [5, 7]. Se-depletion changed expression of a number
of non-selenoprotein genes and pathways, notably the arachidonic acid metabolism pathway. Previous work has indicated
that Se-antioxidant systems influence regulation of arachidonic acid metabolism [49] and can accelerate oxidation of
lipids associated with arachidonic acid metabolism [50]. Furthermore, it has been reported that epoxide hydrolase activity
is affected by Se deficiency [51] and there is experimental evidence for a role of this enzyme activity in arachidonic acid
metabolism [52, 53]. The transcriptomic data showed that expression of EPXH2, which codes for epoxide hydrolase, was
regulated by Se depletion in Caco-2 cells. In addition, EPHX2
was predicted to be a target of the Se-sensitive miR-532-5p and
interestingly epoxide hydrolase activity has been shown previously to be affected by Se deficiency in rats [51]. The observed
change in expression of components of the arachidonic acid
pathway is consistent with reports of Se deficiency in the
mouse affecting expression of genes in the tumour necrosis factor pathway [5] and of GPX2 affecting prostaglandin
metabolism in intestinal cells [19].
In conclusion, the present work demonstrates that alteration in supply of a specific micronutrient can alter the
miRNA expression profile in a human cell line. Specifically,
the data show that Se supply alters expression of 12 miRNA.
We hypothesise that changes in selenoprotein activity lead
to unidentified biochemical events which then in turn regulate miRNA expression. In the case of miR-185, the data
suggest that it is involved in feedback control of expression
of SEPSH2, and possibly also GPX2. This novel regulation
may explain the inverse correlation of expression of these
two selenoproteins with Se intake. Since SEPSH2 is crucial
in Sec-tRNA synthesis such a regulatory mechanism may be
important in maintaining selenoprotein synthesis. Further
work is needed to test this hypothesis, e.g. by using miR-185
mimics and reporter gene assays. It will also be important to
explore whether comparable changes occur in the colon and
if they are related to protective effects of dietary Se against
inflammation and carcinogenesis. Sub-optimal micronutrient intake is regarded as common in populations across the
world, especially in older age [1, 54] and the present data suggest that it will be important to investigate how such changes
in intake affect the impact of miRNA on health-related biochemical pathways.

C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

AMD was supported by the National Council of Science and

Technology (CONACYT). Work in JEH laboratory is supported
by funding from Biotechnology and Biological Sciences Research
Council (UK).
The authors have declared no conflicts of interest.

References

[1] McCann, J. C., Ames, B. N., Adaptive dysfunction of selenoproteins from the perspective of the triage theory: why modest selenium deficiency may increase risk of diseases of aging. FASEB J. 2011, 25, 17931814.
[2] Rayman, M. P., Selenium and human health. Lancet 2012,
379, 12561268.
[3] Prasad, A. S., Discovery of human zinc deficiency: 50 years
later. J. Trace Element Med. Biol. 2012, 26, 6669.
[4] Sinha, A., Cheetham, T. D., Pearce, S. H., Prevention and
treatment of vitamin D deficiency. Calc. Tissue Int. 2012, 92,
207215.

[5] Kipp, A., Banning, A., van Schothorst, E. M., Meplan,

C.
et al., Four selenoproteins, protein biosynthesis, and Wnt
signalling are particularly sensitive to limited selenium
intake in mouse colon. Mol. Nutr. Food Res. 2009, 53,
15611572.
[6] Carlberg, C., Seuter, S., A genomic perspective on vitamin
D signaling. Anticancer Res. 2009, 29, 34853493.
[7] Sunde, R. A., Raines, A. M., Selenium regulation of the
selenoprotein and nonselenoprotein transcriptomes in rodents. Adv. Nutr. (Bethesda, Md.) 2011, 2, 138150.
[8] Cousins, R. J., Blanchard, R. K., Popp, M. P., Liu, L. et al., A
global view of the selectivity of zinc deprivation and excess
on genes expressed in human THP-1 mononuclear cells.
Proc. Natl. Acad. Sci. U S A 2003, 100, 69526957.
[9] Subramanyam, D., Blelloch, R., From microRNAs to targets:
pathway discovery in cell fate transitions. Curr. Opin. Genetics Dev. 2011, 21, 498503.
[10] Ryu, M. S., Langkamp-Henken, B., Chang, S. M., Shankar,
M. N. et al., Genomic analysis, cytokine expression, and microRNA profiling reveal biomarkers of human dietary zinc
depletion and homeostasis. Proc. Natl. Acad. Sci. U S A 2011,
108, 2097020975.
[11] Alvarez-Daz, S., Valle, N., Ferrer-Mayorga, G., Lombarda, L.
et al., MicroRNA-22 is induced by vitamin D and contributes
to its antiproliferative, antimigratory and gene regulatory
effects in colon cancer cells. Hum. Mol. Genet. 2012, 21,
21572165.
[12] Peters, U., Takata, Y., Selenium and the prevention of
prostate and colorectal cancer. Mol. Nutr. Food Res. 2008,
52, 12611272.
[13] Meplan, C., Hesketh, J., The influence of selenium and selenoprotein gene variants on colorectal cancer risk. Mutagenesis 2012, 27, 177186.
[14] Duffield-Lillico, A. J., Reid, M. E., Turnbull, B. W., Combs,
G. F., Jr. et al., Baseline characteristics and the effect of
www.mnf-journal.com

2204

A. Maciel-Dominguez et al.

Mol. Nutr. Food Res. 2013, 57, 21952205

selenium supplementation on cancer incidence in a randomized clinical trial: a summary report of the Nutritional
Prevention of Cancer Trial. Cancer Epidemiol. Biomark. Prevent. 2002, 11, 630639.

[30] Pagmantidis, V., Bermano, G., Villette, S., Broom, I. et al., Effects of Se-depletion on glutathione peroxidase and selenoprotein W gene expression in the colon. FEBS Lett. 2005,
579, 792796.

[15] Chu, F. F., Esworthy, R. S., Doroshow, J. H., Role of Sedependent glutathione peroxidases in gastrointestinal inflammation and cancer. Free Radical Biol. Med. 2004, 36,
14811495.

[31] Bandres, E., Cubedo, E., Agirre, X., Malumbres, R. et al., Identification by Real-time PCR of 13 mature microRNAs differentially expressed in colorectal cancer and non-tumoral tissues. Molecular Cancer 2006, 5, 29, doi: 10.1186/1476-45985-29.

[16] Banning, A., Florian, S., Deubel, S., Thalmann, S. et al.,

GPx2 counteracts PGE2 production by dampening COX-2
and mPGES-1 expression in human colon cancer cells. Antioxid. Redox. Signal 2008, 10, 14911500.
[17] Barnett, M., Bermingham, E., McNabb, W., Bassett, S. et al.,
Investigating micronutrients and epigenetic mechanisms in
relation to inflammatory bowel disease. Mutation Res. 2010,
690, 7180.
[18] Bellinger, F. P., Raman, A. V., Reeves, M. A., Berry, M. J.,
Regulation and function of selenoproteins in human disease.
Biochem. J. 2009, 422, 1122.
[19] Fairweather-Tait, S. J., Bao, Y., Broadley, M. R., Collings, R.
et al., Selenium in human health and disease. Antioxid. Redox. Signal 2011, 14, 13371383.
[20] Berry, M. J., Insights into the hierarchy of selenium incorporation. Nat. Genet. 2005, 37, 11621163.
[21] Kim, J. Y., Carlson, B. A., Xu, X. M., Zeng, Y. et al., Inhibition of
selenocysteine tRNA[Ser]Sec aminoacylation provides evidence that aminoacylation is required for regulatory methylation of this tRNA. Biochem. Biophys. Res. Commun. 2011,
409, 814819.
[22] Carlson, B. A., Schweizer, U., Perella, C., Shrimali, R. K. et al.,
The selenocysteine tRNA STAF-binding region is essential
for adequate selenocysteine tRNA status, selenoprotein expression and early age survival of mice. Biochem. J. 2009,
418, 6171.

[23] Pagmantidis, V., Meplan,

C., van Schothorst, E. M., Keijer, J.
et al., Supplementation of healthy volunteers with nutritionally relevant amounts of selenium increases the expression
of lymphocyte protein biosynthesis genes. Am. J. Clin. Nutr.
2008, 87, 181189.
[24] Kim, V. N., Han, J., Siomi, M. C., Biogenesis of small RNAs
in animals. Nat. Rev. Mol. Cell Biol. 2009, 10, 126139.
[25] van den Berg, A., Mols, J., Han, J., RISC-target interaction:
cleavage and translational suppression. Biochimica et Biophysica Acta 2008, 1779, 668677.
[26] Yeung, M. L., Bennasser, Y., Le, S. Y., Jeang, K. T., siRNA,
miRNA and HIV: promises and challenges. Cell Res 2005, 15,
935946.
[27] Juhila, J., Sipila, T., Icay, K., Nicorici, D. et al., MicroRNA expression profiling reveals miRNA families regulating specific
biological pathways in mouse frontal cortex and hippocampus. PLoS One 2011, 6, e21495.

[32] Chang, K. H., Mestdagh, P., Vandesompele, J., Kerin, M. J.

et al., MicroRNA expression profiling to identify and validate reference genes for relative quantification in colorectal
cancer. BMC Cancer 2010, 10, 173, doi: 10.1186/1471-240710-173.
[33] Kasaikina, M. V., Kravtsova, M. A., Lee, B. C., Seravalli, J.
et al., Dietary selenium affects host selenoproteome expression by influencing the gut microbiota. FASEB J. 2011, 25,
24922499.
[34] Stratmann, J., Wang, C. J., Gnosa, S., Wallin, A. et al., Dicer
and miRNA in relation to clinicopathological variables in
colorectal cancer patients. BMC Cancer 2011, 11, 345, doi:
10.1186/1471-2407-11-345.
[35] Li, J., Zhang, Y., Zhao, J., Kong, F., Chen, Y., Overexpression of miR-22 reverses paclitaxel-induced chemoresistance
through activation of PTEN signaling in p53-mutated colon
cancer cells. Mol. Cell Biochem. 2011, 357, 3138.
[36] Chang, K. H., Miller, N., Kheirelseid, E. A., Lemetre, C. et al.,
MicroRNA signature analysis in colorectal cancer: identification of expression profiles in stage II tumors associated
with aggressive disease. Int. J. Colorectal Dis. 2011, 26,
14151422.
[37] Wang, Y. X., Zhang, X. Y., Zhang, B. F., Yang, C. Q. et al.,
Initial study of microRNA expression profiles of colonic cancer without lymph node metastasis. J. Digest. Dis. 2010, 11,
5054.
[38] Tchernitsa, O., Kasajima, A., Schafer, R., Kuban, R.-J. et al.,
Systematic evaluation of the miRNA-ome and its downstream effects on mRNA expression identifies gastric cancer
progression. J. Pathol. 2010, 222, 310319.
[39] Tsujiura, M., Ichikawa, D., Komatsu, S., Shiozaki, A. et al.,
Circulating microRNAs in plasma of patients with gastric
cancers. Br. J. Cancer 2010, 102, 11741179.
[40] Li, J., Zhang, Y., Zhao, J., Kong, F., Chen, Y., Overexpression of miR-22 reverses paclitaxel-induced chemoresistance
through activation of PTEN signaling in p53-mutated colon
cancer cells. Mol. Cell Biochem. 2011, 357, 3138.
[41] Akcakaya, P., Ekelund, S., Kolosenko, I., Caramuta, S. et al.,
miR-185 and miR-133b deregulation is associated with overall survival and metastasis in colorectal cancer. Int. J. Oncol.
2011, 39, 311318.

[28] Kuhn, D. E., Martin, M. M., Feldman, D. S., Terry, A. V., Jr.
et al., Experimental validation of miRNA targets. Methods
2008, 44, 4754.

[42] Dewing, A. S., Rueli, R. H., Robles, M. J., Nguyen-Wu, E. D.

et al., Expression and regulation of mouse selenoprotein P
transcript variants differing in non-coding RNA. RNA Biol.
2012, 9, 13611369.

[29] Huang, Y., Zou, Q., Song, H., Song, F. et al., A study of miRNA
targets prediction and experimental validation. Protein Cell
2011, 1, 979986.

[43] Brigelius-Flohe, R., Kipp, A., Glutathione peroxidases in different stages of carcinogenesis. Biochimica et Biophysica
Acta 2009, 1790, 15551568.


C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mnf-journal.com

Mol. Nutr. Food Res. 2013, 57, 21952205

[44] Kryukov, G. V., Castellano, S., Novoselov, S. V., Lobanov, A.

V. et al., Characterization of mammalian selenoproteomes.
Science 2003, 300, 14391443.
[45] Xu, Y., Fang, F., Zhang, J., Josson, S. et al., miR-17*
suppresses tumorigenicity of prostate cancer by inhibiting mitochondrial antioxidant enzymes. PLoS One 2010, 5,
e14356.
[46] Moriarty, P. M., Reddy, C. C., Maquat, L. E., Selenium deficiency reduces the abundance of mRNA for Se-dependent
glutathione peroxidase 1 by a UGA-dependent mechanism
likely to be nonsense codon-mediated decay of cytoplasmic
mRNA. Mol. Cell. Biol. 1998, 18, 29322939.
[47] Sunde, R., in: Hatfield, D. L., Berry, M. J., Gladyshev, V. N.
(Eds.), Selenium, Springer, New York 2012, 137152.
[48] Yagublu, V., Arthur, J. R., Babayeva, S. N., Nicol, F.
et al., Expression of selenium-containing proteins in human colon carcinoma tissue. Anticancer Res. 2011, 31,
26932698.
[49] Kurosawa, T., Nakamura, H., Yamaura, E., Fujino, H. et al., Cytotoxicity induced by inhibition of thioredoxin reductases via
multiple signaling pathways: role of cytosolic phospholipase


C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2205
A(2)alpha-dependent and -independent release of arachidonic acid. J. Cell. Physiol. 2009, 219, 606616.
[50] Rock, C., Moos, P. J., Selenoprotein P protects cells from
lipid hydroperoxides generated by 15-LOX-1. Prostaglandins
Leukot. Essent. Fatty Acids 2010, 83, 203210.
[51] Olsson, U., Lundgren, B., Segura-Aguilar, J., MessingEriksson, A. et al., Effects of selenium deficiency on
xenobiotic-metabolizing and other enzymes in rat liver. Int.
J. Vit. Nutr. Res. 1993, 63, 3137.
[52] Morisseau, C., Schebb, N. H., Dong, H., Ulu, A. et al., Role
of soluble epoxide hydrolase phosphatase activity in the
metabolism of lysophosphatidic acids. Biochem. Biophys.
Res. Commun. 2012, 419, 796800.
[53] Vainio, P., Gupta, S., Ketola, K., Mirtti, T. et al., Arachidonic
acid pathway members PLA2G7, HPGD, EPHX2, and CYP4F8
identified as putative novel therapeutic targets in prostate
cancer. Am J. Pathol. 2011, 178, 525536.
[54] Berr, C., Richard, M. J., Roussel, A. M., Bonithon-Kopp,
C., Systemic oxidative stress and cognitive performance in
the population-based EVA study. Etude du Vieillissement
Arteriel. Free Radical Biol. Med. 1998, 24, 12021208.

www.mnf-journal.com