Animal (2020), 14:4, pp 745–752 © The Animal Consortium 2019 animal

doi:10.1017/S1751731119002477

Whole-transcriptome profiling of sheep fed with a high

iodine-supplemented diet
M. Iannaccone1* , R. Elgendy2*a , A. Ianni1, C. Martino3, F. Palazzo1, M. Giantin2 ,
L. Grotta1, M. Dacasto2 and G. Martino1†
1
Faculty of Bioscience and Technology for Food, Agriculture, and Environment, University of Teramo, Via R. Balzarini 1, 64100 Teramo, Italy; 2Department of Comparative
Biomedicine and Food Science, University of Padua, Viale dell’Università 16, 35020 Legnaro, Italy; 3Specialist Diagnostic Department, Istituto Zooprofilattico
Sperimentale dell’Abruzzo e del Molise “G. Caporale”, Via Campo Boario, 64100 Teramo (TE), Italy

(Received 2 May 2019; Accepted 16 September 2019; First published online 23 October 2019)

Iodine (I) is a micronutrient that mammals need for proper functionality of thyroid gland since it is the main component of
thyroid hormones. Besides studies that have investigated the role of I in livestock nutrition, it is also important to know the
transcriptomics changes in small ruminants following I supplementation. Therefore, the aim of this study was to investigate the
effects of I on the whole blood transcriptome in sheep. Fifteen lactating cross-bred ewes (3 to 4-year-old, 55 to 65 kg BW) at
their late lactation period were enrolled in this study. At the beginning, all the animals had a 2-week acclimation period where
they were fed with a basal diet which includes an adequate level of I (2 mg I/animal per day) in the form of calcium iodate
(CaI2O6). Then, the ewes were randomly divided into two groups and fed in individual troughs: the control group (n = 5) was
maintained on basal diet and the experimental group (I, n = 10) was fed for 40 days with a diet containing a high I
supplementation (equivalent to 30 mg I/animal per day), in the form of potassium iodide. Whole blood and milk were collected
individually at the beginning (T0) and after the 40 days of supplementation (T40). Iodine quantification was assessed in serum
and milk sample. Microarray gene expression analysis was performed on whole blood and, filtering data using a fold change >2
with an adjusted P < 0.05, we identified 250 differentially expressed genes (DEGs) in the I group (T40 v. T0). Looking for
biological processes associated with our DEGs, we found significant association with cell growth regulation. Thus, our study
unveils the role of I supplementation on gene expression in sheep improving the knowledge about micronutrients in animal
nutrition.

Keywords: iodine supplementation, microarray, gene expression, pathway analysis, sheep

Implication
Iodine is a micronutrient mineral element currently supplied
to animal’s diets. Iodine effects on productive as well as
reproductive performance have been widely investigated,
but little is known about its role on gene expression
regulation in sheep. Our analysis showed which molecular
pathways are affected by I-supplementation and provide
further knowledge for improving the use of iodine in animal
nutrition. Thus, our results can impact on production system
of this small ruminant providing a new source of informa-
tion about growth and productivity.
* These authors contributed equally to this work.
a
Present address: Department of Immunology, Genetics and Pathology, Uppsala
University, 75185 Uppsala, Sweden
†
E-mail: gmartino@unite.it
Introduction
Effects of different supplementation – such as by-products of
agricultural practices and mineral elements – to the animal
diets have generated a wide interest in the past years
(Elgendy et al., 2016 and 2017; Iannaccone et al., 2018
and 2019). Between them, iodine (I) deserves a high interest
because it is an essential trace element for both humans and
animals. It has only one, but vital, role in the body as a con-
stituent of thyroid hormones, and its importance for livestock
has been well documented over the past years (Schone and
Rajendram, 2009). The I content of farm animals’ diet is
important not only for its impact on the animals’ homeostasis
but also on humans who eventually consume the I-containing
food (e.g., milk) obtained from those animals. However, the
majority of animal feed consist of plants that rarely contain
enough trace elements to meet the animals’ needs

745
Iannaccone, Elgendy, Ianni, Martino, Palazzo, Giantin, Grotta, Dacasto and Martino

(McDowell, 2003). That is why supplementing the animals’ Table 1 Ingredients and chemical composition of the formulated diet
diet with an external source of minerals is a common, and of the lactating ewes
generally desirable, practice. Dietary content
Small ruminants comprise big part of the livestock produc-
Item Basal (CTR) Supplemented (I)
tion, with millions being produced every year (FAO, 2015). In
lactating/pregnant sheep, the recommended level of I is
Ingredients of the total diet
0.80 mg/kg of diet, where the maximum tolerable limits (kg of dietary DM)
are up to 50 mg/kg of diet (NRC, 2007). However, even I Hay (g/kg) 800 800
levels as high as 133 mg/kg of diet were previously reported Concentrate (g/kg) 200 200
to be tolerated by sheep (McCauley et al., 1973). Because of Ingredients of the concentrate
this substantial difference between the recommended and (DM basis)
maximum tolerable levels, and due to the indirect potential Barley (g/kg) 59.55 59.55
effects of I on consumers’ health, several legislations and Soybean flour (g/kg) 22.75 22.75
regulations control the I supplementation to animal feeds. Corn (g/kg) 15.00 15.00
In the European Union (EU), the I supplementation to sheep Calcium carbonate (g/kg) 2.00 2.00
feed was regulated by the directive 1459/2005 (European Dicalcium phosphate (g/kg) 0.50 0.50
Mineral-vitamin mixa (g/kg) 0.20 0.20
Commission (EC), 2005), recommending not to exceed
Iodine (calcium iodate) (mg/kg) 4 4
10 mg/kg I in complete feed. In the recent implementing Iodine (KI) (mg/kg) – 56
regulations (2015/861; EC, 2015), the maximum I content Nutrient composition (DM basis)
in feed for milk-producing ruminants (milk intended for Net energy (NEg) (Mcal/kg) 1.02 1.02
human consumption) has been reduced to 5 mg/kg, while Crude protein (g/kg) 19.59 19.59
it remained at 10 mg/kg of complete feed for other ruminants Crude fat (g/kg) 2.34 2.34
categories. Starch (g/kg) 47 47
The European Food Safety Authority (EFSA, https://www. Crude fiber (g/kg) 5.11 5.11
efsa.europa.eu/) provides, based on published scientific data NDF (g/kg) 17.90 17.90
and evidences, independent scientific advice to the EU ADF (g/kg) 6.01 6.01
authorities on the existing and emerging risks regarding food ADL (g/kg) 1.03 1.03
safety. Since its first opinion on the use of I in animal feeds a
Mineral and vitamin premix contained (per kg) 32.00 mega-IU vitamin A, 3.20
(EFSA, 2005), only few studies have investigated the effects mega-IU vitamin D3, 48 mg vitamin E, 4 mg vitamin B1, 166 mg ferrous carbon-
ate, 2.5 mg sodium selenite, 62 mg manganous oxide and 78 mg zinc oxide.
of I feed levels close to or at the maximum authorized levels
on ruminants. Feeding a diet containing I at 10.1 mg/kg DM
for 2 weeks did not result in observable adverse effects on
lactating Holstein cows (Schone et al., 2009). In fattening cat-
Materials and methods
tle, I at 8.3 mg/kg DM increased the thyroid weight but had
no significant effect on the animals’ productive and reproduc- Animals and study design
tive performances (Meyer et al., 2008). Further, I supplemen- Fifteen lactating (from a herd of 300 subjects) cross-bred
tation of prepartum cows at 15 mg/kg of DM increased the I ewes (3- to 4-year-old, 55 to 65 kg BW) at their late lactation
levels in offspring at birth and 24 h postpartum, but did not period (154 ± 5 days) were used in this study. Ewes had a
affect their ability to absorb immunoglobulin G (IgG) from 2-week acclimation period, in which they were kept on a
colostrum (Conneely et al., 2014). On the contrary, high basal diet that consisted of hay ad libitum (2.5 kg/animal
dietary levels of I administered to sheep were associated with per day (~4.5% of BW)) plus 0.5 kg/animal per day of a
reduced lamb serum IgG concentration and reduced absorp- custom-formulated concentrate (Table 1) offered to the
tion efficiency (Boland et al., 2006 and 2005). Rose et al. animals in two parts during the day (0.3 kg in the morning
(2007) suggested 9.9 mg/kg DM as the upper safe limit of and 0.2 kg in the afternoon).
I supplementation to pregnant ewes, based on the criterion The basal diet was formulated to meet the ewes’ require-
of the concentration of IgG in the plasma of lambs at 24 h ments for maintenance and milk production (NRC, 2007),
after birth. with an adequate level of I (2 mg/animal per day) in the form
Although the effects of high dietary I supplementation on of calcium iodate (CaI2O6). After the acclimation period, and
small ruminants’ or their offspring’s performance and pro- using a randomized pretest-posttest control group experi-
duction have been well documented, little is known about mental design, ewes were randomly assigned to two groups.
the I-induced transcriptional changes in those animals. The first one (n = 5) was given the basal diet and served as a
Thus, the possibility to know the biological pathways behind control (group CTR), while the second one (n = 10) was fed
I supplementation will provide new insights about the effects for 40 days with a diet containing a high I supplementation
on growth and productivity traits. Therefore, the objective of (group I, 10 mg/kg of complete feed; equivalent to I at 30 mg
the present study was to investigate the effects of a high /animal per day), in the form of potassium iodide (KI)
dietary I-supplemented diet on the whole-transcriptome – (ARISTON Feed Srl (http://www.mangimiariston.it/)). The
and its functional interpretation – in lactating ewes. ewes were fed in individual troughs and had a continuous

746

access to water. The final experimental setup consisted
of two groups (CTR and I) with two time-points each
(first (T0) and last (T40) day of the 40-day-supplementation
phase).
At the time of performing the present experimental
trial (mid-2015), the I supplementation was established
not to exceed the maximum level of I (10 mg/kg of com-
plete feed; EC 1459/2005, 2005, http://eur-lex.europa.
eu/legal-content/EN/TXT/?uri=CELEX%3A32005R1459).
Following the recent modifications (EC 2015/861, 2015,
http://eurlex.europa.eu/legalcontent/EN/TXT/?uri=uriserv
%3AOJ.L_.2015.137.01.0001.01.ENG), the amount of
supplemental I here used would be considered as twice
the recommended levels for milk-producing ruminants
(i.e., I at 5 mg/kg of complete feed). However, the milk
of our experimental ewes was not meant for human con-
sumption, and the level of I in the supplemented diet was
kept high only to serve the experimental objectives.
Sampling
Milk yield was recorded at 0 and 40 days of the I supplemen-
tation phase, and individual milk samples (100 ml) were
collected at the same time-points to determine the milk’s
chemical composition. To determine the serum-I level, jugu-
lar venous blood samples were collected at 0 and 40 days of
the I supplementation. For the transcriptome analysis, 2.5 ml
of whole-blood was collected, at T0 and T40, in PAXgene
tubes (Qiagen SpA, Milan, MI, Italy) as per the manufac-
turer’s instructions. The blood-containing PAXgene tubes
were first stored at room temperature overnight and then
at −20°C until RNA isolation.
Iodine determination in serum and milk
The quantification of I in milk and serum was performed by
an inductively coupled plasma mass spectrometer (ICP – MS
Agilent 7500, Agilent, Palo Alto, CA, USA). Regarding blood,
200 μl of serum was weighted directly in Pyrex test tubes;
then, 6 ml of ammonia solution (0.25 M) and 2 ml of H2O2
(30%) were added as extraction solvents. After mineraliza-
tion at 170°C for 30 min in a microwave oven operating at
800 W (Mars Express 5, CEM srl, Cologno al Serio, Italy),
samples were diluted with ultrapure water, centrifuged
and filtered following the method of Comandini et al.
(2013). Six I standard certified solutions (0, 5, 10, 25, 50,
100 μg/l) were used to establish the calibration curve.
Milk-I content at the beginning (T0) and the end (T40) of
the I supplementation phase was determined in both groups
according to Fecher et al. (1998) with some modifications.
For each sample, 0.5 g of milk was homogenized with tetra-
methylammonium hydroxide (0.25 M) and 2 ml of deionized
water (30%). Extraction was carried out in a microwave
(800 W) at a temperature of 170°C for 30 min. After cooling,
samples were transferred into a sterile tube and diluted with
distilled water to a final volume of 15 ml. Samples were then
centrifuged at room temperature for 10 min, filtered through
polytetra-fluoroethylene syringe filters (0.45 μm) and stored
at 4°C until analysis. A standard calibration curve was
Iodine effects on gene expression in sheep
created using six calibration points equaling concentrations
of 0, 5, 10, 25, 50, and 100 μg/l of I in tetramethylammonium
hydroxide. Argon gas was used with a flow rate of 1.05 and
0.2 l/min for the carrier and gas formation, respectively.
Iodine content was determined at m/z = 127 and a total
acquisition time of 21 s. Before the sequence analysis,
ICP-MS was auto-tuned by a solution containing 1 ppb of
different metals (Li, Y, Ce, Tl and Co).
Determination of milk composition
Milk samples, collected at 0, 20, 30 and 40 days of I supple-
mentation, were analyzed for their protein, casein, lactose,
fat, urea, pH and total solids content (%) using Fourier
Transformed InfraRed spectrophotometry (MilkoScan™ FT
6000, Foss, Hillerød, Denmark), as per the manufacturer’s
instructions. The somatic cell count for the milk samples
was analyzed by a fluoro-opto-electronic counter
(Fossomatic™ FC, Foss, Hillerød, Denmark). The method is
based on flow cytometry technology that counts somatic
cells, following the mixing of milk samples (1 to 10 μl, at
30°C to 42°C) with a DNA molecule-binding staining
solution. Milk samples were also evaluated for their total
bacterial count using another type of fluoro-opto-electronic
counter (Bactoscan™ FC, Foss, Hillerød, Denmark). This
automated counter can give an accurate counting of bacteria
in raw milk samples (4.5 ml, at 2°C to 42°C) following an
incubation period that breaks down all milk components
but bacterial cells. A bacteria control sample, a particle con-
trol sample and a frozen milk standard were used to ensure
performance, accuracy and stability of the instrument.
RNA isolation and microarray hybridization
Total RNA was isolated using the PAXgene blood RNA kit
(PreAnalytics/Qiagen, Milan, Italy) as per the manufacturer’s
instructions. Total RNA concentration was measured by
the NanoDrop ND-1000 spectrophotometer (NanoDrop
Technologies Inc., Wilmington, DE, USA). Ribonucleic acid
purity was estimated from the OD 260/280 and the 260/
230 ratios. The RNA integrity was assessed using an
Agilent 2100 Bioanalyzer, using the RNA 6000 Nano kit
(Agilent Technologies, Santa Clara, CA, USA). All samples
with an RNA concentration ≥ 40 ng/μl and an RNA integrity
number ≥ 8 were considered for the subsequent microarray
hybridization. A previously designed and qPCR-validated
custom sheep whole-transcriptome microarray containing
more than 43 000 probes (Elgendy et al., 2016 and 2017)
was used in this study. (The custom-designed microarray
can be found under the accession GPL20576 (https://www.
ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GPL20576).) All
labeling and hybridization procedures were performed as
previously described (Elgendy et al., 2016 and 2017).
Microarray data analyses
Microarray data extraction, filtering and analysis were
performed as previously described (Elgendy et al., 2016).
Briefly, data were normalized by the quantile method
using the limma package (Ritchie et al., 2015) from
747
Iannaccone, Elgendy, Ianni, Martino, Palazzo, Giantin, Grotta, Dacasto and Martino

R/BioConductor. Differentially expressed genes (DEGs) were (a)

identified using the two-class paired option (t-test statistics) 250
**** I
in the Significance Analysis of Microarray software (Tusher

Serum iodine (mg/L)

200 CTR
et al., 2001), applying a false discovery rate (FDR; adjusted
P-value) ≤ 0.05 and a fold-change (FC) > 2. Whole-blood tran- 150
scriptomic I data were analyzed by a repeated-measures
ANOVA followed by Bonferroni’s multiple comparison post- 100
hoc test – using the GraphPad Prism 5 software (San Diego,
CA, USA). Data are reported as means ± SD, and statistical 50
significance was declared at P ≤ 0.05.
To identify enriched biological themes (particularly gene 0
0 40
ontology (GO) terms), and to discover enriched functional- Days
related gene groups (i.e., clusters) in our DEGs, reference (b)
has been made to the Database for Annotation, Visualization 2500
and Integrated Discovery (DAVID, http://david.abcc.ncifcrf. I
gov/) (Huang et al., 2009a and 2009b). The analyses in **** CTR
2000

Milk iodine (mg/L)

DAVID were performed using the default parameters. To visu-
alize the gene expression data of the I group samples, prin- 1500
cipal component analysis (PCA) and hierarchical clustering
were carried out using the web-based ClustVis (http://biit. 1000
cs.ut.ee/clustvis/) tool (Metsalu and Vilo, 2015).
500

Statistics
GraphPad Prism (GraphPad Software, La Jolla, CA, USA) was
used for statistical analysis. Differences in both sera and milk
I content as well as milk composition were all assessed using
ordinary two-way ANOVA without post-test correction. Only
differences with a P < 0.05 were considered statistically
significant.
0
0 40
Days
Figure 1 Iodine quantification in sera and milk of lactating ewes. Iodine
concentration was measured in serum (a) and milk (b) samples from control
(CTR, n = 5) and iodine (I, n = 10) groups at the beginning (T0) and at the
end (T40) of the supplementation period. Data are shown as mean ± SD,
and differences were evaluated using 2-way ANOVA, ****P < 0.0001.

v. 14 ± 13.8 μg/l, P < 0.0001, 2-way ANOVA) at T40 compared

Results
Iodine concentration in serum and milk
All animals remained generally healthy throughout the
experiment, and no toxicity signs were ever noticed among
those fed with the high I-supplemented diet and no
differences were revealed among the groups in terms of
intake (feed intake and refusal, data not shown). The effect
of I supplementation on serum-I concentrations is shown in
Figure 1a. In the I group, the serum-I concentration increased
(P < 0.0001) from 80.4 ± 16.3 μg/l (day 0, baseline) to
208.1 ± 19.5 μg/l at days 40 of I supplementation. In the
CTR group, the serum-I level remained mostly unchanged
during the supplementation period (78.4 ± 18.7 and
72.2 ± 12.4 μg/l at 0 and 40 days, respectively).
The mean (±SD) milk-I content in the CTR and I-supplemented
group at T0 and T40 is reported in Figure 1b. At T0, the milk-I
content was almost similar in both groups (20.6 ± 7.02
v. 14 ± 13.8 μg/l for CTR and I, respectively); at T40, in the
animals belonging to the I group, I amount was ~16 times
higher in comparison to the CTR group (1 421.7 ± 656.7
v. 88.8 ± 30.2 μg/l, P < 0.0001, 2-way ANOVA). Further,
difference in milk-I content in the same group was almost
similar in the CTR group (20.6 ± 7.02 v. 88.8 ± 30.2 μg/l,
P: n.s.) while being about 100 times higher (1 421.7 ± 656.7
with T0 in the I group.
Milk yield and milk composition pattern in experimental
group are reported in Supplementary Material Figure S1.
While the milk production, protein, casein, fat and lactose
content did not differ between the CTR and I-supplemented
groups, an increase (P < 0.001, 2-way ANOVA) in the urea
level was evident at T40 compared with T0, in both groups.
This might indicate a lactation time- and not supplementation-
influenced increase in urea levels.
Influence of iodine supplementation on whole blood
transcriptome
The presence of I-induced changes in gene expression
was investigated by testing for the presence of DEGs at
T40 compared with T0 for both I and CTR groups separately
(I T40 v. T0 and CTR T40 v. T0). In this way we assessed the
I effect (I group) and the ‘temporal’ changes in gene expres-
sion (CTR group, if the gene expression of certain gene
changes normally with time and not supplementation-
dependent).
Using an FC >2 and an FDR <0.05 as threshold values, we
did not find DEGs in the CTR group (CTR T40 v. T0), while 254
DGEs were identified in the I group (I T40 v. T0). Specifically,
250 genes were upregulated, and only 4 were down-
regulated (see Supplementary Material Table S1).

748
 Iodine effects on gene expression in sheep

Figure 2 (colour online) Heatmap and Principal Component Analysis (PCA) of the differentially expressed genes (DEGs) in ewes fed with a high I-supplemented
diet for 40 days (T40 v. T0). (a) Heatmap is generated from DEGs using a hierarchical clustering approach (Euclidean distance clustering algorithm). (b) Principal
Component Analysis plot of animals before (T0; green circles, left side) and after (T40; orange squares, right side) the dietary I supplementation.

Consequently, the following analyses were performed only
for the I group. Using the expression values of the DEGs
(normalized microarray intensities), the T40 and T0 samples
of the I group clustered separately on a hierarchical clustering
heatmap (Figure 2a) and on a PCA plot, with the first two
components amounting to 36.3% of the total variation
(Figure 2b).
The set of up- and downregulated genes was also used for
functional annotation analysis. To identify the most signifi-
cant Kyoto Encyclopedia of Genes and Genomes (KEGG)
pathways and functional annotation clusters in high dietary
I-supplemented ewes, we used the DAVID tool. We first ana-
lyzed the DEG separately, and the four downregulated genes
were not associated with any enrichment annotation, most

749
Iannaccone, Elgendy, Ianni, Martino, Palazzo, Giantin, Grotta, Dacasto and Martino

Table 2 KEGG pathways enriched in the iodine-supplemented lactating ewes group. All list of the enriched KEGG pathways generated using all
upregulated genes following 40-day I-supplemented diet. Pathways are listed by P-value significance, and only the top four have a P < 0.05
Kegg pathway Count P-values Enrichment Genes

oas04510:Focal adhesion 6 0.02 3.7 MYLPF, TGA5, PGF, PRKCG, TLN2

oas04014:Ras signaling pathway 6 0.02 3.5 SYNGAP1, PRKCG, PGF, MAPK10, FGFR4, ENSOARG00000003286
oas04960:Aldosterone-regulated 3 0.032 10.4 NEDD4L, PRKCG, IRS1
sodium reabsorption
oas04726:Serotonergic synapse 4 0.0447 5 ALOX12, PRKCG, RAPGEF3, MAOB
oas04015:Rap1 signaling pathway 5 0.0692 3.2 PRKCG, PGF, TLN2, RAPGEF3, FGFR4
oas04530:Tight junction 4 0.078 3.9 OMYHCS, LLGL1, MYLPF, ENSOARG00000000873
oas04310:Wnt signaling pathway 4 0.0798 3.9 PRKCG, DKK2, MAPK10, DKK4
KEGG = Kyoto encyclopaedia of genes and genomes; I = iodine.

Table 3 All enriched biological process generated using the 250 DEGs from the comparison between the beginning (T0) and the final time point (T40)
of the I supplementation in lactating ewes. Only P < 0.05 are considered statistically significant
Biological processes Counts P-values Genes Enrichment

GO:0035987~endodermal cell differentiation 3 0.022 COL7A1, COL12A1, ITGA5 12.7

GO:0001558~regulation of cell growth 3 0.022 IGFBP2, IGFBP3, IGFBP4 10.9
GO:0016477~cell migration 4 0.044 CDC42BPB, FGFR4, JUP, TNS3 5
GO:0043651~linoleic acid metabolic process 2 0.047 ALOX12, ALOXE3 40.6
GO:0019372~lipoxygenase pathway 2 0.057 ALOX12, ALOXE3 33.8
GO:0008284~positive regulation of cell proliferation 6 0.060 NKX3-1, ST8SIA1, ALOX12, 2.8
FGFR4, PGF, TNS3
GO:0030307~positive regulation of cell growth 3 0.068 INO80, ALOX12, NOL8 6.8
GO:0032688~negative regulation of interferon-beta production 2 0.075 TRAIP, TRAF3IP1 25.4
GO:0019369~arachidonic acid metabolic process 2 0.085 ALOX12, ALOXE3 22.6
DEG = differential expressed genes; I = iodine; GO = gene ontology.

found an enrichment for ‘insulin-like growth factor binding

Annotation cluster (most significant)

protein’ and plekstrin homology domain (Figure 3).

Cluster 1: Insulin-like growth factor
binding protein (INTERPRO)
The most enriched functional GO terms identified with the
enrichment analysis, including also their target genes, are
Cluster 2: Plekstrin homology domain reported in Table 3.
(INTERPRO)

Cluster 3: Ankyrin repeat (INTERPRO)

0 0.5 1 1.5 2
Enrichment score
Figure 3 (colour online) The most enriched clusters following 40-day high
dietary I supplementation in lactating ewes. The 250 upregulated genes were used
for enriched in major functional clusters. Only clusters with enrichment score >1
are showed.
probably because of their low number. Then, the analysis of
the 250 upregulated transcripts showed seven enriched
pathways, but only focal adhesion (oas04510), RAS signaling
(oas04014), aldosterone-regulated sodium reabsorption
(oas04960) and serotonergic synapse (oas04726) were sta-
tistically significant (P < 0.05, Table 2).
While looking for functional clustering, we also applied a
high classification stringency using the same P-value, and we
Discussion
2.5 The main aim of this study was to investigate the transcriptional
effects in the whole blood following a 40-day high dietary I
supplementation in lactating ewes. In the present study, we
demonstrate that a high dietary level of I is likely to induce
change in expression of several genes. We decided to use whole
blood because it is an accessible biological matrix that circu-
lates the whole body and reflects the physio-pathological status
of animals. Moreover, in our previous animal nutrition studies,
using whole blood, we were able to detect DEGs using either
micronutrients (Elgendy et al., 2016 and 2017)) or by-products
(Iannaccone et al., 2018 and 2019).
The supplementation of 2 mg I/animal per day is considered
sufficient (NRC, 2007); thus, the control ewes in our experi-
mental plan would be considered either as I-adequate or
non I-deficient. Further, the I-supplemented group – receiving I

750

at 30 mg/animal per day for 40 days – would be considered
as highly I-supplemented. Together with the fact that CTR
group did not show any time-dependent transcriptional
changes suggests that the results presented here would
be considered a direct effect of I supplementation.
Throughout the supplementation period, the serum-I con-
centration increased exponentially in the experimental group
fed with a I-supplemented diet. However, the absence of
data referring to time-points higher than 40 days prevents
us from making other evaluations on I blood kinetics.
Similar to serum, the milk-I content was significantly higher
in I-supplemented animals compared to that in CTR.
However, the overall milk composition in terms of fat, total
proteins, caseins and lactose percentage was not different
between the two groups. The only exception was the
elevated urea content that was higher at T40 in both groups
indicating a lactation time-dependent effect as already
reported in the previous study, rather than a consequence
of I supplementation (Henao-Velásquez et al., 2014).
Almost 70% to 80% of the total body amount of I is located
in the thyroid, where it represents the principal constituent of
thyroid hormones (Zimmermann, 2009). As a consequence, I is
involved in metabolic processes essential for normal growth
and development. Moreover, it contributes to the basal
metabolism regulation (Brent, 2012), thereby directly influenc-
ing BW and energy expenditure (Fox et al., 2008). In this
regard, the 250 (upregulated) DEGs identified in the present
study, albeit in a short-term exposure study, are a valuable
proof-of-concept of the I capability to affect the blood tran-
scriptome of exposed animals. As a further confirmation of
a I-specific transcriptional effect, no DEGs were noticed in
the CTR group. Indeed, it is unlikely that short-term studies
have a marked effect on gene expression in normal/unstressed
animals which is in line with our previous studies on mineral
supplementation (Elgendy et al., 2016 and 2017).
Investigating more in depth our transcriptional results, we
identified the IGF binding proteins (IGFBPs) as the most
enriched (enrichment fold: 2.33) biological cluster. In our
dataset, the IGFBP-2 (FC: 2.83, P-value: 0.023), IGFBP-3
(FC: 2.02, P-value: 0.044), and IGFBP-4 genes (FC: 3, P-value:
0.032) were all upregulated. These genes belong to a super-
family of proteins (IGFBP-1 to 6) which serve as carrier
proteins for the IGF-1 (Hwa et al., 1999). It has been previ-
ously shown that IGF-1 and IGFBP-3 secretion is influenced
positively by the thyroid status; consequently, the I’s level
can affect the animal’s growth by acting on the axis thyroid
IGF-1→IGFBP-3. In agreement with previous reports
where I supplementation resulted in an improved growth
(Zimmermann et al., 2007), our findings lead us to hypoth-
esize that the upregulation of IGFBP3 is correlated with I sup-
plementation and might positively affect the animal’s growth
without influencing T3/T4 tyroid hormones secretion ratio in
agreement with the previous study (Norouzian et al., 2009).
Thus, when analyzing our data for enriched biological proc-
esses, the ‘regulation of cell growth (GO:0001558)’ was one
out of the four statistically significant (P < 0.05) enriched
pathways. Worth mentioning is also the fact that the most
Iodine effects on gene expression in sheep
enriched biological process was ‘linoleic acid metabolic
process (GO:0043651)’. In fact, thyroid dysregulation has
been shown to influence the linoleic acid metabolism, and
particularly regulating the concentration of its metabolites
derived by lipoxygenase activity (Yao et al., 2015).
Despite further studies involving more animals will be
needed to validate our findings, our data highlight the poten-
tial influence of I supplementation on pathways related to
growth performance and productive traits, contributing to
the growth of nutrigenomics field in ruminants
Acknowledgements
This study was supported by a grant to G. Martino from the
‘Rural Development Plan 2007–2013, MISURA 1.2.4’ –
Region of Abruzzo, Italy (Caratterizzazione e miglioramento
degli indici salutistici e sicurezza alimentare delle produzioni
lattiero ovine tipiche abruzzesi a marchio di origine). We wish
to thank the ‘Associazione Regionale Allevatori d’Abruzzo’
for the fruitful cooperation.
M. Iannaccone 0000-0002-2283-6613
R. Elgendy 0000-0002-2592-3448
M. Giantin 0000-0003-0015-3800
M. Dacasto 0000-0002-3637-5643
G. Martino 0000-0002-7878-9318
Declaration of interest
All authors declare no conflict of interest.
Ethics statement
The study design was approved by the Teramo University
Institutional Animal Care and Use Committee. Animals
were managed according to the Directive 2010/63/EU of the
European Parliament (European Union., 2010), and the
Directive 86/609/EEC (European Economic Community, 1986)
regarding the protection of animals used for experimentation
or other scientific purposes.
Software and data repository resources
The raw and processed microarray data have been deposited
in the Gene Expression Omnibus (GEO) and are accessible
through the accession number GSE74778.
Supplementary material
To view supplementary material for this article, please visit
https://doi.org/10.1017/S1751731119002477
References
Boland T, Callan J, Brophy P, Quinn P and Crosby T 2006. Lamb serum vitamin E
and immunoglobulin G concentrations in response to various maternal mineral
and iodine supplementation regimens. Animal Science 82, 319–325.
Boland T, Guinan M, Brophy P, Callan J, Quinn P, Nowakowski P and Crosby T
2005. The effect of varying levels of mineral and iodine supplementation to ewes
during late pregnancy on serum immunoglobulin G concentrations in their
progeny. Animal Science 80, 209–218.
751
Iannaccone, Elgendy, Ianni, Martino, Palazzo, Giantin, Grotta, Dacasto and Martino
Brent GA 2012. Mechanisms of thyroid hormone action. Journal of Clinical
Investigation 122, 3035–3043.
Comandini P, Cerretani L, Rinaldi M, Cichelli, A and Chiavaro E 2013. Stability of
iodine during cooking: investigation on biofortified and not fortified vegetables
International Journal of Food Sciences and Nutrition 64, 857–861.
Conneely M, Berry DP, Sayers R, Murphy, JP, Doherty ML, Lorenz I and Kennedy E
2014. Does iodine supplementation of the prepartum dairy cow diet affect serum
immunoglobulin G concentration, iodine, and health status of the calf ? Journal
of Dairy Science 97, 1–11.
Elgendy R, Giantin M, Castellani F, Grotta L, Palazzo F, Dacasto M and Martino G
2016. Transcriptomic signature of high dietary organic selenium supplementa-
tion in sheep: a nutrigenomic insight using a custom microarray platform and
gene set enrichment analysis. Journal of Animal Science 94, 3169–3184.
Elgendy R, Palazzo F, Castellani F, Giantin M, Grotta L, Cerretani L, Dacasto M
and Martino G 2017. Transcriptome profiling and functional analysis of sheep
fed with high zinc-supplemented diet: a nutrigenomic approach. Animal Feed
Science and Technology 234, 195–204.
EU Commission 2005. Commission Regulation (EC) No 1459/2005 amending the
conditions for authorization of a number of feed additives belonging to the
group of trace elements. Off. J. Eur. Union.
European Food Safety Authority 2005. Opinion of the scientific panel on addi-
tives and products or substances used in animal feed on the request from the
commission on the use of iodine in feedingstuffs. EFSA-Q-2003-058. The EFSA
Journal 168, 1–42.
European Union (EU) Commission Implementing Regulation 2015/861 of 3 June
2015 concerning the authorisation of potassium iodide, calcium iodate anhy-
drous and coated granulated calcium iodate anhydrous as feed additives for all
animal species. Off. J. Eur. Union.
Fecher P, Goldmann I and Nagengast, A 1998. Determination of iodine in food
samples by inductively coupled plasma mass spectrometry after alkaline extrac-
tion. Journal of Analytical Atomic Spectrometry 13, 977–982.
FAO (Food and Agriculture Organization of the United Nations) 2015. Statistical
pocketbook: World food and agriculture, 2015. Rome, Food and Agriculture
Organization of the United Nations.
Fox C, Pencina M, D’agostino R, Murabito J, Seely E, Pearce E and Vasan R 2008.
Relations of thyroid function to body weight: cross-sectional and longitudinal
observations in a community-based sample. Archives of Internal Medicine
168, 587–592.
Henao-Velásquez AF, Múnera-Bedoya OD, Herrera AC, Agudelo-Trujillo JHand
Cerón-Muñoz MF and 2014. Lactose and milk urea nitrogen: fluctuations during
lactation in Holstein cows. Revista Brasileira de Zootecnia 43, 479–484.
Huang D, Sherman BT and Lempicki RA 2009a. Bioinformatics enrichment tools:
paths toward the comprehensive functional analysis of large gene lists. Nucleic
Acids Research 37, 1–13.
Huang D, Sherman BT and Lempicki RA 2009b. Systematic and integrative analysis
of large gene lists using DAVID bioinformatics resources. Nature Protocol 4, 44–57.
Hwa V, Oh Y and Rosenfeld RG 1999. The insulin-like growth factor-binding
protein (IGFBP) superfamily. Endocrinology Review 20, 761–787.
Iannaccone M, Elgendy R, Giantin M, Martino C, Giansante D, Ianni A,
Dacasto M and Martino G 2018. RNA sequencing-based whole-transcriptome
752
analysis of Friesian cattle fed with grape pomace-supplemented diet. Animals
8, 188.
Iannaccone M, Ianni A, Ramazzotti S, Grotta L, Marone E, Cichelli A and Martino
G. 2019. Whole blood transcriptome analysis reveals positive effects of dried
olive pomace-supplemented diet on inflammation and cholesterol in laying hens.
Animals 9, 427.
McCauley EH, Linn JG and Goodrich RD 1973. Experimentally induced iodide
toxicosis in lambs. American Journal of Veterinary Research 34, 65–70.
McDowell LR 2003. Minerals in animal and human nutrition (No. Ed. 2). Elsevier
Science BV.
Metsalu T and Vilo J 2015. ClustVis: a web tool for visualizing clustering of multi-
variate data using principal component analysis and heatmap. Nucleic Acids
Research 43, W566–570.
Meyer U, Weigel K, Schöne F, Leiterer M and Flachowsky G 2008. Effect of
dietary iodine on growth and iodine status of growing fattening bulls.
Livestock Science 115, 219–225.
National Research Council (NRC) 2007. Nutrient requirements of small rumi-
nants: sheep, goats, cervids, and new world camelids. US National
Academies Press, Washington, DC, USA.
Norouzian MA, Valizadeh R, Azizi F, Hedayati M, Naserian AA and Shahroodi FE
2009. The effect of feeding different levels of potassium iodide on performance,
T3 and T4 concentrations and iodine excretion in Holstein dairy cows. Journal of
Animal and Veterinary Advances 8, 111–114.
Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W and Smyth GK 2015. limma
powers differential expression analyses for RNA-sequencing and microarray
studies. Nucleic Acids Research 43, e47.
Rose MT, Wolf B and Haresign W 2007. Effect of the level of iodine in the diet of
pregnant ewes on the concentration of immunoglobulin G in the plasma of
neonatal lambs following the consumption of colostrum. British Journal of
Nutrition 44, 315–320.
Schone F, Leiterer M, Lebzien P, Bemmann D, Spolders M and Flachowsky G
2009. Iodine concentration of milk in a dose – response study with dairy cows
and implications for consumer iodine intake. Journal of Trace Elements in
Medicine and Biology 23, 84–92.
Schone F and Rajendram R (ed. VR Preedy, GN Burrow and R Watson 2009. Iodine
in farm animals. In Comprehensive handbook of iodine (ed. ), pp. 151–170.
Elsevier, Amsterdam, the Netherlands.
Tusher VG, Tibshirani R and Chu G 2001. Significance analysis of microarrays
applied to the ionizing radiation response. Proceedings of the National
Academy of Sciences of the United States of America 98, 5116–5121.
Yao X, Sa R, Ye C, Zhang D, Zhang S, Xia H, Wang Y, Jiang J, Yin H and Ying H
2015. Effects of thyroid hormone status on metabolic pathways of arachidonic
acid in mice and humans: a targeted metabolomic approach. Prostaglandins &
Other Lipid Mediators 118–119, 11–18.
Zimmermann MB 2009. Iodine deficiency. Endocrinology Review 30, 376–408.
Zimmermann MB, Jooste PL, Mabapa NS, Mbhenyane X, Schoeman S, Biebinger R,
Chaouki N, Bozo M, Grimci L and Bridson J 2007. Treatment of iodine deficiency in
school-age children increases insulin-like growth factor (IGF)-I and IGF binding
protein-3 concentrations and improves somatic growth. The Journal of Clinical
Endocrinology and Metabolism 92, 437–442.