Department of Animal Science, Michigan State University, East Lansing, Michigan 48824
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
differential display; Northern hybridization; spatial expression; fetal development
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
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 (2030%) 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 1020% of fetuses die from day 40 to the end of gestation (29). Critical periods for fetal mortality, coinciding with changes in porcine placental growth, have been found around days 3540, 5575, 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, 4244). 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 insulin-like 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 |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
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 µg) were treated with DNase I (Invitrogen, Carlsbad, CA) in a 10-µl reaction. Synthesis of cDNA consisted of 2.0 µl DNase-treated RNA mixed with 2.0 µl oligo-dT anchor primer (2 µM). The reaction was heated at 70°C for 5 min and quenched on ice. Next, 1x SuperScript buffer, 25 µM 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 µl. 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 µl containing 2 µl cDNA, 1.5 mM MgCl2, 20 µM of each dNTP, 1 U Taq polymerase, 1x MgCl2-free PCR buffer (Promega, Madison, WI), 0.2 µM anchor primer used in the RT reaction, 0.2 µM arbitrary 10-mer sense primer, and 2.5 µCi [-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 24 arbitrary primers) were used corresponding to screening of
10% of all mRNA species present. DDRT-PCR 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 46 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 µl 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 µl 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 µg 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 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 µg) 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 µl 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 µl of 20x SSC and spotted at 1-, 3-, and 5-µg 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 manufacturers directions. Membranes were prehybridized at 65°C for 1 h with 5 ml of hybridization solution (5x SSC, 5x Denhardts 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 2x SSC, 0.1% SDS, twice at 45°C for 10 min with the same solution, twice with 1x SSC, 0.1% SDS for 5 min at 45°C, once with 1x 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 |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
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 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.
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
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 transcripts, 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 alternatively spliced transcripts in the chicken
56 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 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 |
---|
GRANTS
This work was supported by the Michigan Agricultural Experiment Station.
![]() |
FOOTNOTES |
---|
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: ernstc{at}msu.edu).
10.1152/physiolgenomics.00167.2003.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Visit Other APS Journals Online |