Physiol Genomics 16: 268–274, 2004;
10.1152/physiolgenomics.00167.2003.
Developmental changes in the fetal pig transcriptome
Stephanie R. Wesolowski, Nancy E. Raney, and Catherine W. Ernst
Department of Animal Science, Michigan State University, East Lansing, Michigan 48824
Submitted 30 September 2003; accepted in final form 13 November 2003
differential display; Northern hybridization; spatial expression; fetal
development
are complex and the precise control
of gene expression at critical time points is essential for normal
growth and development. It is known that specific genes are
required, yet the overall pattern of gene expression changes
and subsequent signaling pathways is not clearly understood.
Identification of transcriptome changes responsible for the
precise development and growth of specific cells and tissues is
essential to ultimately understand the coordinated events in
fetal development.
The pig is a useful animal model for investigating mechanisms of mammalian genetics and physiology. Understanding
the genes and pathways involved in pig developmental processes will enable more direct efforts toward identifying underlying mechanisms of organ development, fetal growth, and
genetic disorders. During fetal pig development, there is significant embryonic loss (⬃20–30%) before day 20 of gestation,
with the critical period occurring around days 11 and 12 when
peri-implantation conceptuses undergo rapid cellular remodeling (9, 29). Increased fetal growth rates between days 25 and
40 of gestation can lead to uterine crowding and increased fetal
mortality, especially in porcine breeds with high ovulation
rates (35, 36). Another 10–20% of fetuses die from day 40 to
the end of gestation (29). Critical periods for fetal mortality,
DEVELOPMENTAL PROCESSES
Article published online before print. See web site for date of publication
(http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: C. W. Ernst, 1205
Anthony Hall, Dept. of Animal Science, Michigan State Univ., East Lansing,
MI 48824 (E-mail: [email protected]).
268
coinciding with changes in porcine placental growth, have
been found around days 35–40, 55–75, and after day 100 of
gestation (36). Increased number of pigs born alive and rapid
postnatal growth rates are important traits for pig production.
Consequently, understanding the genes involved in normal
fetal development can aid in improving these traits and facilitate more efficient genetic selection programs.
Numerous research efforts have evaluated changes in gene
expression during the early stages of embryonic pig development (22, 26, 37, 42–44). Considerably less is known regarding transcriptome changes during early fetal development despite their importance for fetal survival and growth. Only a
limited number of transcripts have been evaluated after day 20
of fetal development. Relative mRNA abundance of insulinlike growth factor binding proteins (IGFBPs) in skeletal muscle decreased with increasing fetal age, whereas hepatic IGFBP-2 and -5 expression was greatest between days 75 and 89
compared with earlier fetal and later postnatal periods (12).
Also, peak skeletal muscle myostatin expression was identified
around day 50 and sustained throughout the fetal period (15),
and changes in SOX9 and DAX-1 around day 28 were associated with gonadal differentiation (28). Therefore, because
expression patterns of only a few genes have been examined,
additional studies are needed to further characterize transcriptional changes during pig fetal development.
We employed differential display reverse transcription PCR
(DDRT-PCR; 19) to evaluate gene expression differences in
pig fetuses at 21, 35, and 45 days of gestation. Fetuses
experience increased growth rates during these developmental
stages, and knowledge of the genes involved will aid in
understanding the mechanisms controlling this complex process. DDRT-PCR has been utilized to identify differentially
expressed genes involved in growth and developmental processes in multiple species, including human fetal liver and
embryos (21, 25), mouse embryos (46), chicken embryos (8),
and postnatal porcine skeletal muscle (14). Our results provide
new insight regarding transcriptome changes during fetal pig
development that can be used to improve growth and survival
rates to enhance pig production efficiency, as well as for
comparative developmental biology using porcine development as a model for other mammalian species.
MATERIALS AND METHODS
Tissue samples and RNA isolation. Pig fetuses were collected from
Yorkshire ⫻ Landrace crossbred gilts at 21, 35, and 45 days of
gestation (normal gestation is ⬃114 days). To obtain fetuses, gilts
(n ⫽ 2 or 3 per stage) were slaughtered in a federally inspected
abattoir, and fetuses were quickly removed, snap frozen in liquid
nitrogen, and stored at ⫺80°C. Total RNA was extracted from fetal
samples using the RNeasy Maxi Prep Kit (Qiagen, Valencia, CA)
following the manufacturer’s protocol using a maximum of 1 g of
tissue. Three whole 21-day fetuses (from two different litters) weighing ⬃0.5 g each were used for RNA isolation. Fetuses from 35 and 45
days weighed 8–10 g and 20–25 g, respectively. Therefore, to obtain
1094-8341/04 $5.00 Copyright © 2004 the American Physiological Society
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
Wesolowski, Stephanie R., Nancy E. Raney, and Catherine W.
Ernst. Developmental changes in the fetal pig transcriptome. Physiol
Genomics 16: 268–274, 2004; 10.1152/physiolgenomics.00167.2003.—
Growth and development of pig fetuses is dependent on the coordinated
expression of multiple genes. Between 21 and 45 days of gestation,
fetuses experience increasing growth rates that can result in uterine
crowding and increased mortality. We used differential display reverse
transcription-PCR (DDRT-PCR) to identify differentially expressed
genes in pig fetuses at 21, 35, and 45 days of gestation. Pig cDNAs were
identified with homologies to CD3 ␥-subunit, collagen type XIV ␣1,
complement component C6, craniofacial developmental protein 1, crystallin-␥E, DNA binding protein B, ⑀-globin, formin binding protein 2,
ribosomal protein L23, small acidic protein, secreted frizzled related
protein 2, titin, vitamin D binding protein, and two hypothetical protein
products. Two novel expressed sequence tags (ESTs) were also identified. Expression patterns were confirmed for eight genes, and spatiotemporal expression of three genes was evaluated. We identified novel
transcriptome changes in fetal pigs during a period of rapid growth. These
changes involved genes with a spectrum of proposed functions, including
musculoskeletal growth, immune system function, and cellular regulation. This information can ultimately be used to enhance production
efficiency through improved pig growth and survival.
FETAL PIG GENE EXPRESSION
Physiol Genomics • VOL
16 •
gestation fetus was also included to examine hepatic expression of the
identified genes. To evaluate spatial differential expression, Northern
blots were made using total RNA (8 ␮g) from the three 21-day fetal
samples and anterior, medial, and posterior sections from each of the
three 35-day and three 45-day samples, and the 75-day fetal liver
sample. Dot blots contained pooled (within age group) RNA samples
from the 21-, 35-, and 45-day samples, and the 75-day fetal liver
sample. Samples were denatured with 200 ␮l of denaturing solution
(50% formamide, 6% formaldehyde, 0.02 M Tris pH 7.0), incubated
at 65°C for 5 min, and chilled on ice. Samples were combined with
250 ␮l of 20⫻ SSC and spotted at 1-, 3-, and 5-␮g concentrations
onto nylon membranes using the Convertible Filtration Manifold
System (Life Technologies, Grand Island, NY). Membranes were
air-dried and UV cross-linked.
Hybridization and probes. To generate probes for hybridization,
cDNA clones were digested with EcoRI (New England Biolabs,
Beverly, MA), excised from 1% agarose gels, and purified with the
QIAquick Gel Extraction kit (Qiagen). An 18S rRNA clone was used
to verify equality of RNA loading on blots. Gel extracts were labeled
with [␣-32P]dCTP using the Multiprime DNA Labeling System (Amersham Pharmacia Biotech) following the manufacturer’s directions.
Membranes were prehybridized at 65°C for 1 h with 5 ml of hybridization solution (5⫻ SSC, 5⫻ Denhardt’s solution, 0.5% SDS, 0.1%
mg/ml sheared salmon sperm DNA). Fresh hybridization solution and
denatured probe were added and incubated overnight at 65°C. Blots
were washed briefly with 2⫻ SSC, 0.1% SDS, twice at 45°C for 10
min with the same solution, twice with 1⫻ SSC, 0.1% SDS for 5 min
at 45°C, once with 1⫻ SSC, 0.1% SDS for 5 min at 65°C, and
exposed to BioMax MS film (Kodak) until appropriate exposures
were obtained. Signal intensities from autoradiographs were determined using a Gel Doc 2000 Imaging System and the Quantity One
Software (Bio-Rad Laboratories, Hercules, CA). Approximate sizes of
detectable transcripts were calculated relative to the migration distances of 18S and 28S rRNA bands visualized in RNA gels by
ethidium bromide staining prior to transfer.
Statistical analyses. Densitometric values for each hybridization
probe were analyzed by analysis of covariance using the 18S rRNA
values for each blot as a covariate in the generalized linear model
procedure of the Statistical Analysis System (SAS Institute, Cary,
NC). Due to partial confounding of the transcript abundance of SFRP2
and the 18S rRNA abundance, the effects could not be separated by
analysis of covariance, so statistical analysis for this gene was performed by analysis of variance using the ratio of SFRP2 abundance to
18S abundance. Within each analysis, least square means were separated only when the overall P value was significant. Least square
means ⫾ SE of data are presented.
RESULTS
Identification of putative differentially expressed genes.
DDRT-PCR was used to evaluate differences in mRNA transcript abundance during fetal pig development. Using 12
primer pairs, 19 differentially expressed fragments were excised along with two constitutively expressed transcripts (Table 1). To minimize the identification of false positives, three
samples per developmental stage (21, 35, and 45 days of
gestation) were compared. Only bands that displayed consistent patterns within a developmental group and differential
expression patterns between at least one of the other groups
were excised from the gel (Fig. 1). Putative differentially and
constitutively expressed cDNAs were reamplified, cloned, and
sequenced (Table 1). Multiple clones for selected bands were
sequenced to determine whether more than one unique DNA
fragment had been excised from the DDRT-PCR gel, and two
of the selected bands revealed the presence of two distinct
sequences. Also, multiple fragments were identified for two
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
homogeneous, representative fetal tissue samples for RNA isolation
and allow for spatial expression comparisons, three 35-day-old (from
3 different litters) and three 45-day-old (from 3 different litters)
fetuses were divided into anterior, medial, and posterior sections.
Anterior samples consisted primarily of craniofacial, neural, and
forelimb tissues; medial samples consisted of abdomen, spine, and
developing organs; and posterior samples consisted of lower abdominal organs and hindlimb muscular and skeletal tissues. These tissue
sections were powdered with liquid nitrogen using a mortar and
pestle, and 1-g samples of each section were used for RNA extractions. Integrity of each RNA sample was verified by gel electrophoresis and quantity was determined spectrophotometrically.
Differential display reverse transcription-PCR. DDRT-PCR experiments were performed using modifications of published procedures
(19). RNA samples from three 21-day-old whole fetuses and pools of
equal quantities of anterior, medial, and posterior section samples for
each of three 35-day-old and three 45-day-old fetuses were used.
Oligonucleotide primers were provided by the US Pig Genome
Coordination Program. RNA samples (1.0 ␮g) were treated with
DNase I (Invitrogen, Carlsbad, CA) in a 10-␮l reaction. Synthesis of
cDNA consisted of 2.0 ␮l DNase-treated RNA mixed with 2.0 ␮l
oligo-dT anchor primer (2 ␮M). The reaction was heated at 70°C for
5 min and quenched on ice. Next, 1⫻ SuperScript buffer, 25 ␮M of
each dNTP, 10 mM DTT, and 40 U SuperScript II (Invitrogen) were
added to the RNA and primer solution in a final volume of 20 ␮l. This
mixture was incubated at 40°C for 5 min, 50°C for 5 min, 70°C for 15
min, and held at 4°C. Subsequent PCR reactions were performed in a
total volume of 20 ␮l containing 2 ␮l cDNA, 1.5 mM MgCl2, 20 ␮M
of each dNTP, 1 U Taq polymerase, 1⫻ MgCl2-free PCR buffer
(Promega, Madison, WI), 0.2 ␮M anchor primer used in the RT
reaction, 0.2 ␮M arbitrary 10-mer sense primer, and 2.5 ␮Ci
[␣-33P]dATP. PCR conditions began with denaturation at 95°C for 2
min, four cycles at 92°C for 15 s, 50°C for 30 s, and 70°C for 2 min,
followed by an additional 25 cycles with annealing at 60°C and a final
extension at 72°C for 2 min. A total of 12 primer pairs (3 anchor
primers each paired with 2–4 arbitrary primers) were used corresponding to screening of ⬃10% of all mRNA species present. DDRTPCR samples were mixed with loading dye (formamide, bromphenol
blue, xylene cyanol), denatured at 95°C for 3 min, chilled on ice, and
loaded onto 0.4-mm 5.2% polyacrylamide denaturing gels. Gels were
run at 60 W for 4–6 h. After transfer to filter paper and drying under
a vacuum at 80°C, gels were exposed to BioMax MR film (Kodak,
Rochester, NY) overnight. Fragments that amplified in all three
samples of at least one developmental age group and were faint or
undetectable in the remaining age(s) were excised, placed in microcentrifuge tubes with 100 ␮l of double-distilled H2O, heated at 50°C
for 30 min or at 37°C for 1 h, and stored at ⫺20°C.
Cloning and sequencing. DDRT-PCR products were reamplified
using 2.0 ␮l of excised gel band and the PCR conditions described
above, except isotope was omitted and either the M13 reverse and T7
primers (the 5⬘ ends of anchor and arbitrary primers contained M13
and T7 sequences, respectively) or the original differential display
reaction primers were used. An aliquot of the PCR was analyzed on
a 1% agarose gel with a ␭ HindIII marker (Invitrogen) to estimate
concentration. PCR products were cloned using the pGEM-T Easy
Vector System (Promega). Plasmid DNA was purified using a QIAquick Plasmid Mini kit (Qiagen) and sequenced using SP6 or M13
forward primers with an ABI 373 Dye Terminator cycle sequencing
kit (PerkinElmer Applied Biosystems, Foster City, CA). DNA sequence identities were determined using the basic local alignment
search tool (BLAST) software and the nonredundant database of
GenBank.
Northern and dot blots. For Northern blots, 8 ␮g of total RNA from
the nine samples used in the DDRT-PCR reactions were electrophoresed in 1.2% agarose formaldehyde gels, transferred to Hybond
nylon membranes (Amersham Pharmacia Biotech, Piscataway, NJ),
and UV cross-linked. A liver RNA sample derived from a 75-day
269
270
FETAL PIG GENE EXPRESSION
Table 1. Summary of DDRT-PCR clone sequence identities and gene expression validation results
Length,
bp
Accession
Number
Sequence Identity
Sequence Homology
0901E1
0913B2
0901F1
0901D3
0702F2
0702C1
0702B1
0702A1
0716H1
0904A6
0702J1
0716G1
0901G2
0901D2
0901C2
0702E1
0901F2
0913H4
0901I2
583
314
311
121
146
214
413
719
185
780
235
396
270
119
182
148
312
159
300
BE693215
BE693212
BE693211
BE644600
BE693214
BE693221
BE644605
BE693220
BE693218
BE644601
BE693222
BE693213
BE693219
BE644603
BE644602
BE644604
BE693216
BE693217
BE693223
CD3 ␥-subunit (CD3G)
Collagen type XIV ␣1 (COL14A1)
Complement component C6 (C6)
Craniofacial developmental protein 1 (CFDP1)
Crystallin; ␥-E (CRYGE)
DNA binding protein B (DBPB)
ε-Globin (EG)*
Formin binding protein 2 (FNBP2)
Ribosomal protein L23 (RPL23)*
Small acidic protein (SMAP)
Secreted frizzled related protein 2 (SFRP2)
Titin (TTN)
Vitamin D binding protein (DBP)
DKFZP161131514
FLJ10298
No significant match
No significant match
Activator of heat shock protein (AHSAI)
Tetraspan NET-2 (NET-2)
X52994, 2e⫺63 (O)
XM_044622, e⫺109
NM_000065, 2e⫺12
NM_006324, 4e⫺27
NM_173289, 2e⫺20 (R)
M95794, 7e⫺52 (B)
X86792, e⫺131 (P)
XM_059095, e⫺132
BC010114, 5e⫺43
BC020937, 0.0
XM_050625, 9e⫺64
AJ560658, e⫺162 (P)
BC022310, 2e⫺18
NM_032288, e⫺8
BC032998, e⫺15
BC007398, 8e⫺23
AF124522, 2e⫺71
DDRT-PCR
Gel
Northern, Dot
Blot
Increase
Increase
Increase
Decrease
Increase
Decrease
Decrease
Increase
Decrease
Decrease
Increase
Increase
Increase
Decrease
Decrease
Decrease
Increase
Constitutive
Constitutive
NA
Increase,NS
Increase
NA
NA
NA
Decrease
NS
Decrease
Decrease
Increase
Increase,NS
Increase
NA
NA
NA
Increase,NS
NA
NA
Primers used for differential display reverse transcription PCR (DDRT-PCR) were obtained from the US Swine Genome Coordinator, and primer sequences
are available at http://www.genome.iastate.edu/resources/ddprimer.html. Primers used to obtain each clone are indicated by the first four digits of the clone
name (1st and 2nd digits, anchor primer; 3rd and 4th digits, arbitrary primer). Sequence homology indicates GenBank accession number and expectation value
for most significant match. Matches are relative to human unless followed by a letter code in parentheses (B, bovine; O, ovine; P, porcine; R, rat). For
“DDRT-PCR gel” and “Northern Blot, Dot Blot,” relative pattern of mRNA abundance across developmental ages is indicated: “increase” indicates gene
expression increased as fetal age increased; “decrease” indicates expression decreased with fetal age; “constitutive” indicates similar expression across all age
groups; “NS” indicates expression validation was performed, but statistically significant differential expression was not confirmed; “NA” indicates no expression
validation was performed and/or expression was not detected with Northern or dot blot analyses. *Multiple fragments were identified for the gene.
genes. In total, 19 distinct sequences (17 differentially and 2
constitutively expressed) were found from the 21 cloned products.
Putative differentially expressed cDNA fragments were
found with significant sequence homologies to CD3 ␥-subunit
(CD3G), collagen type XIV ␣1 (COL14A1), complement component C6 (C6), craniofacial developmental protein 1
(CFDP1), crystallin-␥E (CRYGE), DNA binding protein B
(DBPB), ⑀-globin (EG), formin binding protein 2 (FNBP2),
ribosomal protein L23 (RPL23), small acidic protein (SMAP),
secreted frizzled related protein 2 (SFRP2), titin (TTN), and
vitamin D binding protein (DBP). Two clones had homology
with the hypothetical protein products DKFZP161131514 and
FLJ10298. Additionally, two sequences failed to reveal homology to any known genes. Sequencing of the two constitutively
expressed fragments revealed homology to activator of heat
shock protein (AHSA1) and tetraspan NET-2 (NET-2), respectively.
Confirmation of differential expression. To confirm differential expression of selected genes, a combination of Northern
and dot blot analyses were utilized. Vitamin D binding protein,
titin, and clone 0901F2 showed increasing relative mRNA
Fig. 1. Differential display of fetal pig cDNA. RNA isolated from pig fetuses
at 21, 35, and 45 days of gestation (three per stage) was used for differential
display reverse transcription PCR (DDRT-PCR). Putative differentially expressed cDNA is indicated by the arrow.
Physiol Genomics • VOL
16 •
abundance, and small acidic protein showed decreasing expression with increasing developmental stage by dot blot analyses
(Fig. 2). Northern blot analyses for TTN and clone 0901F2
revealed the presence of multiple, large transcripts which
compromised quantitative comparisons among individual animals within the developmental age groups. Consequently, this
must be considered when evaluating dot blot results for these
two genes.
Fig. 2. Confirmation of differential expression by dot blotting. Pools of 21, 35,
and 45 days of gestation fetal RNA samples and a 75-day fetal liver (L) RNA
sample were spotted at 1, 3, and 5 ␮g. Shown are representative autoradiographs of dot blots probed with vitamin D binding protein (DBP), small acidic
protein (SMAP), titin (TTN), and novel expressed sequence tag (EST) 0901F2.
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
Clone
Name
FETAL PIG GENE EXPRESSION
group for this transcript. Ribosomal protein L23 analyses
showed decreased relative mRNA abundance in 45-day animals compared with the earlier developmental stages, whereas
SFRP2 expression was lower at 21 days compared with the
later stages, with the exception of one 45-day fetal sample that
exhibited low expression. Small acidic protein relative mRNA
abundance was higher at 21 days compared with 35 and 45
days. Relatively abundant expression of C6, DBP, and the
smaller COL14A1 transcript was detected in the 75-day fetal
liver sample. In addition, expression of RPL23 and SMAP was
found in fetal liver, although the transcript abundance was less
than in the whole fetal samples.
To evaluate spatial expression, Northern blots containing
total RNA from anterior, medial, and posterior sections of
35-day and 45-day fetuses were utilized. Relative mRNA
abundance of DBP was high in 35- and 45-day posterior
samples and undetectable in anterior samples (Fig. 4). A
Fig. 3. Relative abundance of mRNAs during fetal development. Total cellular RNA (8 ␮g per lane) was used in Northern blot
analyses to confirm the expression of collagen type 14 ␣1 (COL14A1), complement component C6 (C6), DBP, ⑀-globin (EG),
formin binding protein 2 (FNBP2), ribosomal protein L23 (RPL23), secreted frizzled related protein 2 (SFRP2), and SMAP. A
representative autoradiograph of a Northern blot probed with each respective cDNA and the same blot probed with 18S rRNA is
presented. Each blot contained the nine fetal samples (21, 35, and 45 days of gestation, n ⫽ 3 per age) used in DDRT-PCR and
a 75-day fetal liver sample. Relative mRNA abundance was quantified and analyzed using 18S rRNA data for each transcript as
a covariate. For SFRP2, data were analyzed by analysis of variance using the ratio of SFRP2 abundance to 18S rRNA abundance.
Least square means ⫾ SE are graphically presented. Bars with different letters are significant (P ⬍ 0.05). Approximate transcript
sizes are indicated to the left of the blots.
Physiol Genomics • VOL
16 •
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
Semi-quantitative Northern blot analyses were conducted for
eight genes (Fig. 3). Other genes listed in Table 1 were
subjected to Northern hybridizations; however, no transcript
signals were detected. Northern blot analyses of COL14A1
revealed two detectable transcripts at ⬃5.8 and 1.6 kb. Only
the longer transcript is presented showing a trend for increasing
expression with increasing age. Complement component C6
and DBP both had less relative mRNA abundance at 21 days
compared with 35 and 45 days. ⑀-Globin Northern analysis
revealed two transcripts at ⬃3.2 and 1.2 kb. The shorter, more
abundant transcript had greatest expression in 21-day samples
compared with 35- and 45-day samples. A similar pattern of
differential expression was observed for the longer, less abundant transcript (data not shown). Formin binding protein 2
Northern analyses failed to confirm any significant changes in
mRNA expression, which may be explained by the considerable variation observed among individual animals within each
271
272
FETAL PIG GENE EXPRESSION
similar spatiotemporal expression pattern was detected for EG;
however, relative transcript abundance in the 35- and 45-day
sections was substantially less than in the whole 21-day fetuses. Variability in expression was observed for both the DBP
and EG transcripts within medial samples from individual
animals in each developmental stage. Expression of these
genes in the 21-day whole fetal samples compared with the 35and 45-day developmental stages was consistent with the
results presented in Fig. 3. SFRP2 revealed detectable transcripts in all spatial and temporal samples with variable relative
mRNA abundance.
DISCUSSION
Our results provide new information regarding changes in
the fetal pig transcriptome (i.e., the collection of all expressed
mRNA transcripts) during development. We identified 17 differentially expressed genes including four with sequence homology to hypothetical or novel genes. Identification of these
novel transcripts is one advantage of the DDRT-PCR technique
(32), and future investigations into the function of these uncharacterized genes is of interest to advance our understanding
of fetal development.
We verified the expression patterns of 10 genes (⬃60%)
using Northern and dot blot analyses, and significant differential expression was confirmed for most of these genes. This
high confirmation rate may be attributed to many factors. We
used multiple RNA samples per developmental stage, which
were isolated using Qiagen columns that yield low amounts of
small RNA molecules and contaminating genomic DNA. Only
bands that were differentially expressed among all the samples
within an age group were extracted, to reduce identification of
false positives (32). It has been suggested that amplification of
multiple transcripts which comigrate in the gels may be a
source of false positives in DDRT-PCR (24). In recognition of
this, we sequenced multiple clones of isolated fragments and
found only two fragments that contained two unique tranPhysiol Genomics • VOL
16 •
scripts, which differed in size by one or two base pairs and
would thus be indistinguishable in the gels (CFDP1 and
0901D2 from the same fragment; C6 and 0901F2 from the
same fragment). Northern analyses were attempted for CFDP1
and 0901D2 but transcripts were undetectable. Differential
expression of C6 was confirmed by Northern analysis. Expressed sequence tag (EST) 0901F2 revealed multiple transcripts by Northern analysis and showed differential expression
by the dot blot method. This particular transcript also failed to
reveal homology with any known sequences but did have short
sequence similarities with a repeat element. The amplification
of DDRT-PCR fragments containing repetitive elements is
another complication of the DDRT-PCR technique. In our
experiments, 0901F2 was the only fragment that possessed
homology to any repeat element.
We detected multiple transcripts for COL14A1, EG, TTN,
and 0901F2. Since we utilized RNA from whole fetal tissue
samples at each developmental stage, the existence of multiple
cell and tissue types presents the possibility to detect all
transcripts and isoforms being expressed at that particular time
point. Titin is the largest known polypeptide and a major
myofibrillar component of muscle with multiple isoforms in
cardiac and striated muscle (40). It has been proposed that the
proteins from these splice variants work together to give
muscle its elastic abilities (41). ⑀-Globin is part of the large
globin locus consisting of five homologous members. The
epsilon gene is specifically expressed during the embryonic
stage followed by fetal and adult globins later in development
(3, 5). The smaller transcript (⬃1.2 kb) is more abundant and
likely represents the ⑀-globin gene based on transcript size
comparison with the human gene (1). The faint larger transcript
(⬃3.2 kb) is present only in day 21 tissue samples and may
represent transcript(s) for other globin gene family members.
Our COL14A1 Northern hybridization probe detected two
transcripts at ⬃5.8 kb and 1.6 kb. Only the smaller transcript
was detectable in the 75-day fetal liver sample. This gene has
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 4. Spatiotemporal relative mRNA abundance evaluated by Northern blot analyses. Total RNA (8 ␮g) for each of three 21-day
fetuses and anterior (A), medial (M), and posterior (P) samples for each of three 35-day and three 45-day fetuses was used to
evaluate spatial differential expression. A 75-day fetal liver sample was also included. Representative autoradiographs for DBP,
EG, SFRP2, and 18S rRNA are shown.
FETAL PIG GENE EXPRESSION
Physiol Genomics • VOL
16 •
T-cell antigen receptor complex (34). Both transcripts had
increased expression later in development potentially corresponding to more advanced immune systems. Increased understanding of the immune system of fetal pigs is important for
potential use of porcine organs for human transplants.
⑀-Globin is part of the hemoglobin system and is normally
expressed in the embryonic yolk sac. Two ⑀-chains in combination with two ␨-chains form embryonic hemoglobin Hb
Gower I, and two ⑀-chains with two ␣-chains constitute embryonic Hb Gower II (3, 5). The observed decrease in expression of EG suggests that between 21 and 35 days of fetal age
in the pig, there is a switch from ⑀-globin expression to the
other fetal hemoglobin forms.
Additional isolated transcripts could be involved in regulating cellular and signaling processes. Secreted frizzled related
protein 2 is a member of the frizzled transmembrane protein
family, which are receptors for Wnt pathway signaling molecules. Secreted frizzled related proteins appear to compete with
membrane-bound frizzled receptors for binding of secreted
Wnt ligands to modulate Wnt signaling (24). Chang et al. (4)
found mRNA expression of this gene exclusively in adult
bovine retina, suggesting that it may be involved in determining the polarity of photoreceptors. However, our spatial relative mRNA abundance results suggest that expression of this
gene is not limited to the lens, as it is expressed in all fetal
sections. Thus SFRP2 may have additional functional roles in
fetal pig development. DNA binding protein B, also known as
nuclease-sensitive element binding protein 1 (NSEP1) and
y-box transcription factor (YB1), has been shown to have
greatest expression in human skeletal muscle and heart and is
a transcriptional regulator (17). Ribosomal protein L23 belongs
to the L14P family of ribosomal proteins and is localized in the
cytoplasm. Expression has been found in all human tissues
examined (2, 13). Formin binding protein 2 has been shown to
be overexpressed in cancer cell lines. Recent characterization
of this gene identified multiple conserved protein domains and
expression in tumorigenic tissues (16). Consequently, this
transcript may be involved in cell cycle control or mitogenic
pathways during fetal development. The functional properties
of SMAP are currently unknown.
This work reveals novel developmental changes in the fetal
pig transcriptome and extends previously described functions
for several of these genes in fetal development. Of future
interest is to further elucidate the molecular mechanisms of
these genes and their respective protein products in specific
signaling and functional pathways, especially for the novel
ESTs. Results of this research increase our knowledge regarding fetal growth and developmental processes and can be used
to aid in understanding normal growth and the consequences of
molecular disorders in the pig and other mammalian species.
ACKNOWLEDGMENTS
We thank Dr. Max Rothschild, U.S. Swine Genome Coordinator, for
distribution of the DDRT-PCR oligonucleotide primers. We are grateful to Dr.
Robert Tempelman for assistance with statistical analyses.
GRANTS
This work was supported by the Michigan Agricultural Experiment Station.
REFERENCES
1. Baralle FE, Shoulders CC, and Proudfoot NJ. The primary structure of
the human epsilon-globin gene. Cell 21: 621–626, 1980.
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
alternatively spliced transcripts in the chicken ⬃5–6 kb in size
(10, 11, 39). Consequently, the small transcript we are detecting could be a different gene with homology to COL14A1 or
a unique and novel porcine alternative splice variant.
Our spatiotemporal Northern blot analyses provide a novel
qualitative assessment regarding the general localization of
transcripts during fetal pig development. Vitamin D binding
protein and EG were localized toward the posterior side of the
fetuses and were undetectable on the anterior side. The medial
and posterior samples consisted mostly of abdominal organ
tissues and muscular and skeletal structural tissues. Differential
spatiotemporal expression of DBP and EG suggests that these
genes play integral roles in growth and development of these
tissues. The detectable, yet variable, expression of SFRP2 in all
spatial samples suggests that this gene functions in multiple
tissues throughout the fetus.
Functional experiments are needed to adequately describe
the roles of the differentially expressed genes identified in this
experiment. However, previous studies in other species provide
clues to possible functional roles for the coordinated expression of these transcripts during fetal development. Three of the
transcripts we isolated are known to be involved in developing
fetal muscle and skeletal tissues. Collagen 14A1, also known
as undulin, is a component of the extracellular matrix in
connective tissue and collagen fibrils (30). It is expressed in
early embryonic chicken tissues and later in collagen containing tissues, including skeletal and cardiac muscle, tendon,
periosteum, and nerve tissue (39, 45). Vitamin D binding
protein is a multifunctional protein found in plasma, ascetic
fluid, cerebrospinal fluid, and many cell surfaces. It binds
vitamin D and its metabolites and transports them to target
tissues, in particular bone (6, 38). Titin, as previously described, is a major gene in muscle tissue. We observed an
increase in COL14A1, DBP, and TTN abundance during the
fetal period, which likely coincides with increasing skeletal
growth and myogenesis during this period.
Craniofacial developmental protein 1 and CRYGE have
been shown to be involved in cranial and facial developmental
processes. First discovered in the mouse as a novel gene
(cp27), CFDP1 is involved in tooth development and has since
been shown to be expressed in various mouse embryonic
tissues and mouse postnatal developing teeth, heart, lung, and
liver (7). The bovine ortholog, BCNT, contains a LINE repetitive insert (27), which is present in this gene in ruminant
species but not in human or pig (33). Luan and Diekwisch (20)
showed that CFDP1 is involved in the viability and proliferation of mouse embryonic fibroblasts and proposed that this
gene may be involved in organogenesis. In our experiment,
CFDP1 expression was greatest in early development, suggesting a role in earlier developmental processes. Crystallin-␥E is
part of the ␥-crystallin multigene family, which is expressed
only in the eye lens. Siezen et al. (31) found expression of the
crystallins in young human lenses. We observed increasing
expression of this gene with increasing developmental stage
which could correlate with eye formation and development.
Complement C6 and CD3G are integral components of
distinct immune systems. C6 is involved in the lytic pathway,
or common final pathway, of the complement system cascade.
C6 and the remaining late complement components (C5
through C9) form membrane attack complexes in an effort to
lyse bacteria or viruses (18). CD3G is a component of the
273
274
FETAL PIG GENE EXPRESSION
Physiol Genomics • VOL
16 •
24. Miele G, MacRae L, McBride D, Manson J, and Clinton M. Elimination of false positives generated through PCR re-amplification of differential display cDNA. Biotechniques 25: 138–144, 1998.
25. Monk M, Holding C, and Goto T. Isolation of novel developmental
genes from human germ cell, oocyte and embryo cDNA by differential
display. Reprod Fertil Dev 13: 51–57, 2001.
26. Murnane RD, Wright RW Jr, Hoffman KA, and Prieur DJ. Expression of beta-galactosidase in preimplantation ovine and porcine embryos.
Proc Soc Exp Biol Med 194: 144–148, 1990.
27. Nobukuni T, Kobayashi M, Omori A, Ichinose S, Iwanaga T, Takahashi I, Hashimoto K, Hattori S, Kaibuchi K, Miyata Y, Masui T, and
Iwashita S. An Alu-linked repetitive sequence corresponding to 280
amino acids is expressed in a novel bovine protein, but not in its human
homologue. J Biol Chem 272: 2801–2807, 1997.
28. Parma P, Pailhoux E, and Cotinot C. Reverse transcription-polymerase
chain reaction analysis of genes involved in gonadal differentiation in
pigs. Biol Reprod 61: 741–748, 1999.
29. Pope WF. Embryonic mortality in swine. In: Embryonic Mortality in
Domestic Species, edited by Geisert RD. Boca Raton: CRC, 1994, p.
53–78.
30. Schuppan D, Cantaluppi MC, Becker J, Veit A, Bunte T, Troyer D,
Schuppan F, Schmid M, Ackermann R, and Hahn EG. Undulin, an
extracellular matrix glycoprotein associated with collagen fibrils. J Biol
Chem 265: 8823–8832, 1990.
31. Siezen RJ, Thomson JA, Kaplan ED, and Benedek GB. Human lens
gamma-crystallins: isolation, identification, and characterization of the
expressed gene products. Proc Natl Acad Sci USA 84: 6088–6092, 1987.
32. Stein J and Liang P. Differential display technology: a general guide.
Cell Mol Life Sci 59: 1235–1240, 2002.
33. Takahashi I, Nobukuni T, Ohmori H, Kobayashi M, Tanaka S,
Ohshima K, Okada N, Masui T, Hashimoto K, and Iwashita S.
Existence of a bovine LINE repetitive insert that appears in the cDNA of
bovine protein BCNT in ruminant, but not in human, genomes. Gene 211:
387–394, 1998.
34. Tunnacliffe A, Buluwela L, and Rabbitts TH. Physical linkage of three
CD3 genes on human chromosome 11. EMBO J 6: 2953–2957, 1987.
35. Van der Lende T and Schoenmaker GJW. The relationship between
ovulation rate and litter size before and after day 45 of pregnancy in gilts
and sows. Livest Prod Sci 26: 217–229, 1990.
36. Van der Lende T and Van Rens BT. Critical periods for foetal mortality
in gilts identified by analysing the length distribution of mummified
foetuses and frequency of non-fresh stillborn piglets. Anim Reprod Sci 75:
141–150, 2003.
37. Vaughan TJ, James PS, Pascall JC, and Brown KD. Expression of the
genes for TGF␣, EGF, and the EGF receptor during early pig development. Development 116: 663–669, 1992.
38. Verboven C, Rabijns A, De Maeyer M, Van Baelen H, Bouillon R, and
De Ranter C. A structural basis for the unique binding features of the
human vitamin D-binding protein. Nat Struct Biol 9: 131–136, 2002.
39. Walchli C, Koch M, Chiquet M, Odermatt BF, and Trueb B. Tissuespecific expression of the fibril-associated collagens XII and XIV. J Cell
Sci 107: 669–681, 1994.
40. Wang K, McClure J, and Tu A. Titin: major myofibrillar components of
striated muscle. Proc Natl Acad Sci USA 76: 3698–702, 1979.
41. Wang K, McCarter R, Wright J, Beverly J, and Ramirez-Mitchell R.
Regulation of skeletal muscle stiffness and elasticity by titin isoforms: a
test of the segmental extension model of resting tension. Proc Natl Acad
Sci USA 88: 7101–7105, 1991.
42. Wilson ME, Sonstegard TS, Smith TPL, Fahrenkrug SC, and Ford
SP. Differential expression during elongation in the preimplantation pig
embryo. Genesis 26: 9–14, 2000.
43. Yelich JV, Pomp D, and Geisert RD. Detection of transcripts for retinoic
acid receptors, retinal-binding protein, and transforming growth factors
during rapid trophoblastic elongation in the porcine conceptus. Biol
Reprod 57: 286–294, 1997.
44. Yelich JV, Pomp D, and Geisert RD. Ontogeny of elongation and gene
expression in the early developing porcine conceptus. Biol Reprod 57:
1256–1265, 1997.
45. Young BB, Gordon MK, and Birk DE. Expression of type XIV collagen
in developing chicken tendons: association with assembly and growth of
collagen fibrils. Dev Dyn 217: 430–439, 2000.
46. Zimmerman JW and Schultz RM. Analysis of gene expression in the
preimplantation mouse embryo. Proc Natl Acad Sci USA 91: 5456–5460,
1994.
www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.2 on June 17, 2017
2. Berchtold MW and Berger MC. Isolation and analysis of a human
cDNA highly homologous to the yeast gene encoding L17A ribosomal
protein. Gene 102: 283–288, 1991.
3. Brotherton TW, Chui DH, Gauldie J, and Patterson M. Hemoglobin
ontogeny during normal mouse fetal development. Proc Natl Acad Sci
USA 76: 2853–2857, 1979.
4. Chang JT, Esumi N, Moore K, Li Y, Zhang S, Chew C, Goodman B,
Rattner A, Moody S, Stetten G, Campochiaro PA, and Zack DJ.
Cloning and characterization of a secreted frizzled-related protein that is
expressed by the retinal pigment epithelium. Hum Mol Genet 8: 575–583,
1999.
5. Choi OR and Engel JD. Developmental regulation of beta-globin gene
switching. Cell 55: 17–26, 1988.
6. Cooke NE and David EV. Serum vitamin D-binding protein is a third
member of the albumin and alpha fetoprotein gene family. J Clin Invest
76: 2420–2424, 1985.
7. Diekwisch TG, Marches F, Williams A, and Luan X. Cloning, gene
expression, and characterization of CP27, a novel gene in mouse embryogenesis. Gene 235: 19–30, 1999.
8. Dietz UH, Ziegelmeier G, Bittner K, and Balling R. Spatio-temporal
distribution of chondromodulin-I mRNA in the chicken embryo: expression during cartilage development and formation of the heart and eye. Dev
Dyn 216: 233–243, 1999.
9. Geisert RD, Brookbank JW, Roberts RM, and Bazer FW. Establishment of pregnancy in the pig: cellular remodeling of the porcine blastocyte
during elongation on day 12 of pregnancy. Biol Reprod 27: 941–955,
1982.
10. Gerecke DR, Foley JW, Castagnola P, Gennari M, Dublet B, Cancedda R, Linsenmayer TF, van der Rest M, Olsen BR, and Gordon
MK. Type XIV collagen is encoded by alternative transcripts with distinct
5⬘ regions and is a multidomain protein with homologies to von Willebrand’s factor, fibronectin, and other matrix proteins. J Biol Chem 268:
12177–12184, 1993.
11. Gerecke DR, Meng X, Liu B, and Birk DE. Complete primary structure
and genomic organization of the mouse Col14a1 gene. Matrix Biol 22:
209–216, 2003.
12. Gerrard DE, Okamura CS, and Grant AL. Expression and location of
IGF binding proteins-2, -4, and -5 in developing fetal tissues. J Anim Sci
77: 1431–1441, 1999.
13. Herault Y, Michel D, Chatelain G, and Brun G. cDNA and predicted
amino acid sequences of the human ribosomal protein genes rpS12 and
rpL17. Nucleic Acids Res 19: 4001, 1991.
14. Janzen MA, Kuhlers DL, Jungs SB, and Louis CF. ARP-16 is upregulated in the longissimus muscle of pigs possessing an elevated growth
rate. J Anim Sci 78: 1475–1484, 2000.
15. Ji S, Losinski RL, Cornelius SG, Frank GR, Willis GM, Gerrard DE,
Depreux FF, and Spurlock ME. Myostatin expression in porcine tissues:
tissue specificity and developmental and postnatal regulation. Am J
Physiol Regul Integr Comp Physiol 275: R1265–R1273, 1998.
16. Katoh M and Katoh M. FNBP2 gene on human chromosome 1q32.1
encodes ARHGAP family protein with FCH, FBH, RhoGAP and SH3
domains. Int J Mol Med 11: 791–797, 2003.
17. Kudo S, Mattei MG, and Fukuda M. Characterization of the gene for
dbpA, a family member of the nucleic-acid-binding proteins containing a
cold-shock domain. Eur J Biochem 231: 72–82, 1995.
18. Law SKA and Reid KBM. Complement (2nd ed.). Oxford: IRL, 1995.
19. Liang P and Pardee AB. Differential display of eukaryotic messenger
RNA by means of the polymerase chain reaction. Science 257: 967–971,
1992.
20. Luan X and Diekwisch TG. CP27 affects viability, proliferation, attachment and gene expression in embryonic fibroblasts. Cell Prolif 35: 207–
219, 2002.
21. Malhorta K, Luehrsen KR, Costello LL, Raich TJ, Sim K, Foltz L,
Davidson S, Xu H, Chen A, Yamanishi DT, Lindemann GW, Cain CA,
Madiansacay MR, Hashima SM, Pham TL, Mahoney W, and
Schueler PA. Identification of differentially expressed mRNAs in human
fetal liver across gestation. Nucleic Acids Res 27: 839–847, 1999.
22. Meijer HA, Van De Pavert SA, Stroband HW, and Boerjan ML.
Expression of the organizer specific homeobox gene goosecoid (gsc) in
porcine embryos. Mol Reprod Dev 55: 1–7, 2000.
23. Melkonyan HS, Chang WC, Shapiro JP, Mahadevappa M, Fitzpatrick PA, Kiefer MC, Tomei LD, and Umansky SR. SARPs: a family
of secreted apoptosis-related proteins. Proc Natl Acad Sci USA 94:
13636–13641, 1997.