Introduction
Litter size traits are the most important traits for reproduction to
increase economic yields with regard to the breeding of pigs. The
improvement of these traits through conventional methods revealed a low rate
of efficiency due to low levels of heritability (Okere and Nelson, 2002;
Grandinson et al., 2003). In fact, fertility traits are affected by season,
parity and breed as well as housing and feeding conditions (Knecht et al.,
2015). The usefulness of molecular genetics may generate a marker to improve
the reproductive performance traits in pigs. Currently, both genomic and
transcriptomic (microarray and RNA sequencing) approaches have been used to
analyze quantitative trait loci (QTL) and to assess candidate genes for
reproductive traits in pigs. Several genes have been discovered. One of
these genes is osteopontin (OPN) which can be differentially expressed in
various reproductive tissues, e.g., ovarian follicles, ovaries, myometrium and
endometrium (Caetano et al., 2004; Fernandez-Roddriguez et al., 2011; Sun et
al., 2011; Wang et al., 2013; Samborski et al., 2013).
Primer sequence, PCR product size and restriction enzymes
of SNPs markers of the porcine OPN gene.
SNP ID
SNP position
Location
Primer sequence
PCR product
Restriction
size (bp)
enzyme
rs341073067
c.278A>G
Exon 5
F: 5′-GCCATCCACCAATGAGGCTA-3′
189
TaaI
R: 5′-ATGCCGGACCTTGGATTCAG-3′
rs81509049
c.425G>A
Exon 6
F*: 5′-TCCGAGGAAGCTGATCGCG-3′
188
Hin6I
rs81214160
c.452T>G
R*: 5′-GATTTTGACCTCAGTCCGT-3′
HinfI
rs81509053
c.573T>C
Pfl23II
rs81214162
c.881C>T
Exon 7
F : 5′-AGGACAGTCAGGAGACGAG-3′
200
TasI
R: 5′-TTCTTCGCTCTTAGAGTCTG-3′
* Mismatch bases are underlined
The OPN or secreted phosphoprotein 1 (SPP1) gene is an extracellular matrix (ECM)
protein and intergrin-binding ligand, as well as a highly phosphorylated
acidic glycoprotein that stimulates cell–cell adhesion and increases
cell–ECM communication, and promotes cell migration, cell signaling and
remodeling in the uterus and placenta (Garlow et al., 2002; Johnson et al.,
2003; White et al., 2005). The OPN gene is located on the long arm of the Sus scrofa
chromosome 8 (SSC8q) and is composed of seven exons and six introns, which are
encoded with a protein of 303 amino acids (Ensembl database:
http://asia.ensembl.org/Sus_scrofa/Info/ Index). The
expression levels of OPN mRNA and the protein are increased in the uterus of
pigs during pregnancy (Garlow et al., 2002; White et al., 2005). Some
polymorphisms of the porcine OPN gene have been identified within a 5′-flanking
region and the promoter including exon 6 and intron 6 (Muráni
et al., 2009; Goluch et al., 2009; Korwin-Kossakowska et al., 2013).
Additionally, a highly polymorphic site of porcine OPN gene (ENSSSCG00000009216)
has been reported in the Ensembl database (http://asia.ensembl.org/index.html).
A total of 353 single nucleotide polymorphisms (SNPs) of the porcine OPN gene were identified
(consisting of 5 missense, 7 synonymous, 5 3′-UTR, 132 intron, 100
upstream and 104 downstream gene variants; online date 10.04.2015). The
short interspersed nuclear element (SINE) polymorphisms in the intron 6 and microsatellite markers of the porcine
OPN gene have only been used to test for an association with the total number born (TNB), number born alive (NBA) and
the number of piglets weaned (NW) as well as sperm quality traits in pig
breeds (Short et al., 1997; Southwood et al., 1998; Knoll et al., 1999;
Hamann et al., 2000; Putnova et al., 2001; Korwin-Kossakowska et al., 2002;
Lin et al., 2006; Niu et al., 2008). Presently, there has still been
no validation of the SNPs on the coding sequence of the porcine OPN gene
(especial missense or non-synonymous mutation) as well as in terms of their
association with the litter size traits in pigs. The objective of this study
was to elucidate the effects of non-synonymous polymorphisms in the coding
sequence of the porcine OPN gene on litter size traits in Thai commercial breed
pigs.
Material and methods
Animals and DNA extraction
Blood samples were taken from a total of 320 sows of the Thai Large White
pig breed. All animals were obtained from Betagro Hybrid International
Company, Thailand. The reproductive performance traits of the sows were
recorded in terms of litter size traits consisting of total number born
(TNB) and number born alive (NBA). Genomic DNAs were extracted using the
Chelex method (Walsh et al., 1991) and kept at 4 ∘C until
analyzed.
Single nucleotide polymorphisms genotyping
The single nucleotide polymorphisms of the porcine OPN gene were genotyped
using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. The specific primers were designed based on the
porcine OPN nucleotide sequences (GenBank accession number: NM_214023) as
shown in Table 1. Additionally, the mismatched primers were
designed to generate a recognition site of the restriction enzyme for
genotyping (Table 1). These designed primers covered five non-synonymous SNPs
of the porcine OPN gene (c.278, c.425, c.452, c.573 and c.881;
http://asia.ensembl.org/index.html).
The PCR was performed in a final volume of 20 µL containing 50 ng of
genomic DNA sample, 1×NH4SO4 buffer, 1.5 mM MgCl2, 0.2 mM
dNTPs, 0.4 µM of each primer and 0.2 U Taq DNA polymerase (Fermentas). The
PCR conditions were 94 ∘C for 3 min denaturing, followed by 40 cycles
of 94 ∘C for 30 s denaturing, 55 ∘C for 30 s annealing,
72 ∘C for 30 s extension and then 5 min at 72 ∘C to complete the extension.
The PCR products were digested with 2.5 U of the restriction enzyme
(Fermentas) for each primer and incubated for 2 h (Table 1). The digested
fragments were separated on a 8 % polyacrylamide gel electrophoresis in
1×TBE buffer and stained with ethidium bromide.
Statistical analysis
The genotype and allele frequencies were calculated for all SNPs. The
haplotype and linkage disequilibrium (LD) of each SNPs in the porcine OPN gene
were analyzed using Haploview software version 4.2 with the solid spine of the LD
method (Barrett et al., 2005). Association analysis of the OPN gene and the
litter size traits was performed using a general linear model (GLM) of the
SAS software package version 9.0 (SAS Institute Inc., Cary, NC, USA),
including the fixed effects of season of the year, parity, genotype and residual
error. The following model was used:
Yijkl=μ+Si+Pj+Gk+eijkl,
where Yijkl is representative of the observed values of the phenotype
traits, μ represents the average normalized record of the populations,
Si represents the fixed effect of season of the year (i= 1–7), Pj
represents the fixed effect of the parity (j=1 and ≥2), Gk is
the fixed effect of genotype (k= 1–3), and eijkl is representative of the
residual error. The additive effect (a) was estimated by the comparison of
the means of the trait value for a homozygote: a=12 (BB–AA) and
the dominance effect (d) for the alleles A and B was calculated from the
means of the following three genotypes: d= AB -12 (AA+BB).
The estimated effects were evaluated using a t test on the significant
deviation from zero (Lin et al., 2006).
Results
Verification of porcine <italic>OPN</italic> polymorphisms
In order to verify the polymorphisms of the porcine OPN gene, five
non-synonymous SNPs were selected to be tested in the Thai commercial pig
population. Three polymorphic sites of the porcine OPN gene were segregated
among Thai Large White pigs, consisting of c.425G>A,
c.573T>C and c.881C>T (Fig. 1). These polymorphisms
revealed a non-synonymous mutation leading to a non-conservative amino acid
exchange at the position p.110Ala>Thr, p.159Val>Ala
and p.262Pro>Ser, respectively. No polymorphisms of the two SNP
markers (c.278A>G and c.452T>G) were observed in
this study.
Genotyping SNPs of the porcine OPN gene (a) at
c.425G>A locus with Hin6I (b) at c.573T>C locus with
Pfl23II and (c) at c.881C>T locus with TasI. The molecular marker of
100 bp DNA ladder (M) and the OPN genotypes are indicated at the top of each lane.
![]()
Genotype and allele frequencies of porcine <italic>OPN</italic> gene
The genotype and allele frequencies of the porcine OPN gene at each polymorphic
site are shown in Table 2. Three SNPs were found to be segregated among
Thai Large White sows. At the c.425 and c.881 loci, three genotypes were found
to be present, whereas at the c.573 locus, two genotypes were present in this
pig population. Alleles of c.425G, c.573T and c.881C were the major alleles
in this pig breed. Moreover, at the c.278 and c.452 loci, these two SNPs were
fixed as c.278A and c.452T among Thai Large White sows.
Genotype and allele frequencies of the porcine OPN gene in
Thai Large White sows.
Locus
Genotype frequency
Allele frequency
AA
AB
BB
A*
B
c.278A>G
1.00
0.00
0.00
1.00
0.00
c.425G>A
0.19
0.39
0.42
0.38
0.62
c.452T>G
0.00
0.00
1.00
0.00
1.00
c.573T>C
0.87
0.13
0.00
0.93
0.07
c.881C>T
0.69
0.26
0.05
0.82
0.18
* Allele A represents wild type alleles of c.278A, c.425G, c.452T, c.573T and
c.881C for each locus and allele B represents mutate alleles of c.278G,
c.425A, c.452G, c.573C and c.881T for each locus.
Haplotype analysis
Haplotype block and graphical representation of LD structure were generated.
A haplotype block was found in the porcine OPN gene (Fig. 2). There was a
strong linkage disequilibrium (LD) found between the SNPs at position c.425
and c.573 (D′= 0.85). On the other hand, the SNPs c.425 and c.881
or the c.573 and c.881 were found to be in the moderate range of LD
(D′= 0.66, D′= 0.76, respectively) among Thai Large
White pigs.
Linkage disequilibrium (LD) plot of three SNPs (c.425,
c.573 and c.881) of the porcine OPN gene. The values in squares represent the
pairwise D′ values and the haplotype block is indicated by a
stronger LD.
![]()
Association of the porcine OPN gene with litter size traits.
SNPs
Parity
Traits
Genotypes (mean ± SE)
Additive
Dominance
GG
GA
AA
c.425
First parity
TNB
9.77 ± 0.82
10.79 ± 0.52
10.32 ± 0.58
-0.27± 0.49
0.74 ± 0.25
NBA
8.67 ± 0.89
9.40 ± 0.56
9.67 ± 0.63
-0.50± 0.35
0.23 ± 0.11
Later parities
TNB
10.90 ± 0.39a
12.49 ± 0.19b
12.29 ± 0.21b
-0.69± 0.21d
0.89 ± 0.41c
NBA
9.72 ± 0.38a
11.46 ± 0.19b
11.19 ± 0.20b
-0.73± 0.21c
1.00 ± 0.29d
TT
TC
CC
c.573
First parity
TNB
10.51 ± 0.38
10.55 ± 0.94
0
-
-
NBA
9.36 ± 0.48
10.00 ± 1.01
0
-
-
Later parities
TNB
12.43 ± 0.14a
10.26 ± 0.44b
0
-
-
NBA
11.33 ± 0.14a
9.52 ± 0.44b
0
-
-
CC
CT
TT
c.881
First parity
TNB
10.50 ± .45
10.23 ± 0.58
10.99 ± 1.89
-0.24± 0.37
-0.51± 0.12
NBA
9.73 ± 0.47
8.58 ± 0.61
11.34 ± 0.34
-0.80± 0.23
0.95 ± 0.48
Later parities
TNB
12.48 ± 0.16a
11.72 ± 0.24a
10.86 ± 0.95b
0.81 ± 0.44
0.04 ± 0.50
NBA
11.45 ± 0.15a
10.57 ± 0.23a
8.91 ± 0.92b
1.27 ± 0.47c
0.38 ± 0.52
Mean ± SE represents least square mean ± standard error.
a, b Values in each row are
significantly different (P<0.05). c, d Statistically
significant differences of the additive and dominance effects; cP<0.01; dP<0.001.
Association of porcine <italic>OPN</italic> polymorphisms with litter size traits
The results of the association analysis between three SNPs (c.425, c.573 and
c.881) and litter size traits are shown in Table 3. No significant
association of SNPs with litter size traits were observed in the first
parity of pigs. In later parities, these three SNPs were significantly
associated (P<0.05) with the TNB and NBA traits of sows. At the c.425
locus, the sows with the AA and GA genotypes had significantly higher TNB
and NBA values than those of the sows with the GG genotype. At the c.573 locus,
the sows with the TT genotype revealed significantly higher TNB and NBA
values than those of the sows with the TC genotype. At the c.881 locus, the sows
with the CC and CT genotypes had significantly higher TNB and NBA values
than those of the sows with the TT genotype. The significant additive and
dominance effects for TNB and NBA were detected in later parities at the c.425
locus. Moreover, the significant additive effect for NBA was observed in
later parities at the c.881 locus. Effects of genotype combinations of c.425,
c.573 and c.881 on litter size traits are shown in Table 4. No association
of genotype combinations with litter size traits for the first parity were
seen. In later parities, genotype combinations of these SNPs were
significantly associated with TNB and NBA traits in Thai Large White sows.
The number of favorable alleles was related with the TNB and NBA traits.
Association of genotype combinations of SNP c.425, c.573
and c.881 with litter size traits.
Genotype
Parity
Traits
Number of favorable alleles (mean ± SE)
combinations
0
1
2
3
4
c.425-c.573
First parity
TNB
–
10.19 ± 1.23
10.33 ± 0.96
10.56 ± 0.56
10.48 ± 0.62
NBA
–
10.21 ± 1.13
8.37 ± 1.02
9.23 ± 0.60
9.85 ± 0.66
Later parities
TNB
–
10.21 ± 0.57a
11.10 ± 0.44a
12.56 ± 0.19b
12.33 ± 0.21b
NBA
–
9.50 ± 0.56a
9.77 ± 0.44a
11.50 ± 0.19b
11.23 ± 0.21b
c.425-c.881
First parity
TNB
10.98 ± 1.87
9.49 ± 0.93
10.76 ± 0.19
10.74 ± 0.57
10.22 ± 0.67
NBA
11.46 ± 1.99
7.91 ± 0.99
9.25 ± 0.94
9.34 ± 0.60
9.95 ± 0.72
Later parities
TNB
10.81 ± 0.95ab
10.90 ± 0.44a
12.14 ± 0.37b
12.47 ± 0.20b
12.39 ± 0.23b
NBA
8.87 ± 0.93a
9.77 ± 0.43a
11.03 ± 0.36b
11.46 ± 0.19b
11.28 ± 0.22b
c.573-c.881
First parity
TNB
–
11.40 ± 2.16
9.96 ± 1.09
10.65 ± 0.66
10.51 ± 0.46
NBA
–
11.79 ± 2.30
9.22 ± 1.16
8.72 ± 0.70
9.74 ± 0.49
Later parities
TNB
–
11.29 ± 0.23abc
9.91 ± 0.54a
11.90 ± 0.26b
12.58 ± 0.15c
NBA
–
10.34 ± 0.21abc
8.76 ± 0.53a
10.75 ± 0.26b
11.53 ± 0.15c
Mean ± SE represents least square mean ± standard error. Values in each row with different superscript letters are
significantly different (P<0.05). Number of favorable alleles is
accumulated alleles of combination genotypes for c.425A, c.573T and c.881C.
Discussion
The function of the OPN gene is involved in uterine endometrial glandular
epithelium, adhesion and communication between the conceptus trophectoderm
with uterine endometrial luminal epithelium during early implantation
period, influence fetal/placental development, growth and mediate
communication between placental and uterine tissues that support pregnancy
(Garlow et al., 2002; Johnson et al., 2003). Furthermore, the OPN gene
corresponds to the location with the QTL regions for reproductive traits,
such as ovulation rate (Rathje et al., 1997), uterine capacity (Rohrer et
al., 1999), age of puberty (Cassady et al., 2001), litter size and
embryo-survival rate (King et al., 2003), on SSC8 in pigs. Moreover, the
function genomic analysis showed that the differential expression levels of
the OPN gene were found in the ovarian follicles, ovaries, myometrium and
endometrium (Caetano et al., 2004; Fernandez-Roddriguez et al., 2011; Sun et
al., 2011; Wang et al., 2013; Samborski et al., 2013). Therefore the OPN gene
can be regarded as a candidate gene for the determination of reproductive
traits in pigs.
The polymorphisms of the porcine OPN gene have been identified (Knoll et al.,
1999; Zhang et al., 1992). The SINE polymorphisms at intron 6 of the porcine
OPN gene were found to be associated with litter size as well as body weight at
birth and weaning traits (Korwin-Kossakowska et al., 2002; Niu et al., 2008;
Zhang et al., 2010). These SINE polymorphisms were found to be associated
with the uterine weight in Polish Landrace pigs (Kapelanski et al., 2013).
Moreover, two synonymous SNP (c.559A>C and c.574C>G)
of the porcine OPN gene were associated with their expression levels in uterine
tissues (Korwin-Kossakowska et al., 2013). Additionally, the microsatellite
markers of this gene were associated with the TNB and NBA traits in sows
(Short et al., 1997; Southwood et al., 1998) as well as boar fertility
traits (Lin et al., 2006).
Several polymorphisms in the 5′-flanking region and promoter of the
porcine OPN gene have also been analyzed (Muráni et al., 2009; Goluch et
al., 2009; Korwin-Kossakowska et al., 2013). The g.3836A>G locus
regulated the porcine OPN expression and affected the CCAAT/enhancer binding
protein beta (C/EBPβ) responsive transcriptional enhancer (Muráni
et al., 2009). Moreover, two SNPs at positions g.1999A>G
(-617A>G) and g.2011A>G (-606 A>G) loci
of the porcine OPN promoter region were associated with mRNA expression levels
in the uterus (Korwin-Kossakowska et al., 2013).
In this study, three of five non-synonymous SNPs (c.425G>A,
c.573T>C and c.881C>T) of the porcine OPN gene were
found to be segregated in Thai Large White sows. The c.425 and c.573 SNPs
revealed strong linkage disequilibrium. These three SNPs (c.425G>A, c.573T>C and c.881C>T) of the porcine OPN gene
showed a significant association with the TNB and NBA traits. The positive
effects of the favorable c.425A, c.573T and c.881C alleles on TNB and NBA
were found in later parities. The increased number of favorable alleles was
associated with the TNB and NBA traits in sows. Moreover, the high frequency
of the allele c.425A, c.573T and c.881C were observed in this population of
sows. These results indicate a relatively high frequency of these alleles
along with the high selection pressure for litter size traits in this
commercial pig breed. Moreover, two SNPs (g.1999A>G and
g.3836A>G) of the 5′-flanking region and the promoter
were also detected in this study revealing a lower polymorphisms in Thai
Large White pigs (data not shown).
In this study, the association of non-synonymous SNPs at the c.573T>C locus of the porcine OPN gene with litter size traits is of a significant
level of interest due to the changing amino acid of p.159Val>Ala
located on the 154RGDSVVYGLR163 integrin binding sites, which is
bound to integrin receptors on conceptus trophectoderm and uterine luminal
epithelium. It is involved in conceptus elongation and implantation (Erikson
et al., 2009). The c.573T>C polymorphism may be distrusted in
the molecular binding mechanism of the intergrin receptors and their ligand
of the conceptus and uterine to promote trophectoderm cell migration and the
attachment to luminal epithelium. Our results indicate that the
non-synonymous polymorphisms of the porcine OPN gene are significantly
correlated with litter size traits. This SNP may directly play an important
function in the early stage of the development of the term of pregnancy in
pigs. Further studies should be conducted on the effects of these SNPs on
protein function as well as molecular affinity for its target molecules.
In conclusion, we report on the non-synonymous SNPs in the porcine OPN gene and
their associations with litter size traits. The SNPs at c.425G>A, c.573T>C and c.881C>T loci were found to be
associated with the TNB and NBA traits in Thai Large White pigs. Moreover,
the SNPs at c.425 and c.573 were shown to be in strong linkage
disequilibrium. These findings emphasize the importance of the porcine OPN gene
in the reproductive traits of pigs. Therefore, these SNPs of the porcine
OPN gene may be a potential candidate gene for the purposes of increasing
litter size traits in the breeding of pigs.