1 Biotechnology and Germplasm Laboratory, United States Department of Agriculture Agricultural Research Service, Beltsville, Maryland
2 Bovine Functional Genomics Laboratory, United States Department of Agriculture Agricultural Research Service, Beltsville, Maryland
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
preimplantation development; gene regulation; transcript profiles; serial analysis of gene expression
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Porcine conceptuses develop through a series of four critical transitional stages prior to implantation. Development from the two- to four-cell stages, day 2 (D2) after insemination, coincides with the transition from maternal to embryonic genome expression within the conceptus (39). Morphological differentiation and cellular polarization occur within the blastocyst stage conceptus at day 6 (D6) after insemination (26, 32). The blastocyst also represents the stage most commonly transferred or cryopreserved in swine (11). Beginning at D11, porcine conceptuses undergo dramatic elongation from an 810 mm ovoid to a >150-mm filament by D12 (15). The phenomenon of conceptus elongation also occurs in the sheep and cow at D11 and day 13 (D13), respectively (5). This morphological change occurs via hyperplasia in the sheep and cow, whereas in the pig the initial stages of trophectoderm elongation take place through cellular reorganization and differentiation, and proliferation is a later event (15, 30). Asynchrony in the timing of elongation among D11 and D12 swine conceptuses in utero may play an important role in conceptus survival; blastocysts differentiating earlier may have a competitive edge over their lagging cohorts in obtaining the uterine surface necessary for further development (5). Coinciding with elongation, D12 swine conceptuses also synthesize and secrete estrogens that serve as an important molecular signal in establishing the maternal recognition of pregnancy (16, 30).
Spatial and temporal patterns of gene expression are highly orchestrated during preimplantation conceptus development. Studies utilizing semiquantitative PCR analyses, suppressive subtractive hybridization (SSH), RNA arbitrarily primed-PCR (RAP-PCR), or expressed sequence tags (EST) have identified transcripts that have important physiological roles during trophectoderm elongation (34, 35, 4143). For example, transcripts for steroidogenic enzymes CYP19A and 17-hydroxylase (CYP17A1) were upregulated between gestational D11 and D12, a pattern consistent with increased estrogen synthesis (30, 43). Although these methodologies have been instrumental in assessing gene expression, they exhibit one or more limitations that diminish their utility for the establishment of a more comprehensive gene profile of the transcriptome for 1) efficient gene discovery, 2) sensitivity for the detection of rare transcripts, and 3) direct quantitative transcriptome-wide assessment.
Serial analysis of gene expression (SAGE) enables sensitive transcriptome-wide qualitative and quantitative analysis of gene expression within tissues during discrete physiological states (37, 40). Additionally, SAGE does not require extensive knowledge of the genome information or access to cDNA libraries and clones. Our laboratory recently demonstrated the applicability of SAGE in the assessment of differential gene expression in a previously uncharacterized turkey sperm storage tubule model system that lacked publicly available DNA sequence information (22). We employed SAGE to characterize and compare gene expression in in vivo produced pig conceptuses obtained at two critical stages of development: at initiation of signals for maternal recognition of pregnancy (D11 ovoid) and after transformation from ovoid to D12 filamentous in preparation for implantation. The aim was to establish comprehensive gene expression profiles that would indicate genes that are differentially expressed during normal conceptus development and begin to elucidate the metabolic pathways modulated by these genes during elongation of the pig conceptus.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Tissue collection and total RNA preparation.
Two gilts each were slaughtered on the morning of gestational days 11 or 12. Reproductive tracts were excised from the gilts, and conceptuses were flushed into a petri dish using 75 ml of ice-cold PBS into the oviduct of each individual uterine horn. Morphologies of the D11 and D12 conceptuses were ovoid (610 mm) and filamentous (100150 mm), respectively. Approximately five conceptuses that displayed similar morphology were isolated and collected from each uterine horn of the two gilts. Conceptuses derived from the same animal and time point were pooled then placed either in lysis solution (ToTALLY RNA kit; Ambion, Austin, TX), immediately homogenized, and stored at 80°C, or placed in cryovials and snap-frozen in liquid nitrogen for storage. Total RNA was isolated from a pool of ovoid or filamentous conceptuses using the ToTALLY RNA kit (Ambion) according to the manufacturers protocol. Quantification and analysis of the integrity of total RNA were performed with the Agilent 2100 Bioanalyzer and RNA 6000 Nano LabChip Kit (Agilent Technologies, Palo Alto, CA). All animal protocols were approved by the Beltsville Area Animal Care and Use Committee and meet the United States Department of Agriculture and National Institutes of Health guidelines for the care and use of animals.
SAGE library construction and sequencing.
SAGE libraries were constructed from 10 µg total RNA from ovoid or filamentous conceptuses utilizing the I-SAGE kit (Invitrogen, Gaithersburg, MD) according to the manufacturers recommendation (22). The I-SAGE kit is based on the methodology established by Velculescu et al. (40). Amplification of SAGE concatemers and sequence was carried out as previously described (22).
Processing and analysis of SAGE tag sequences.
The quality assessment of sequences was performed using PHRED with a minimum quality score of 18 and minimum sequence length of 50 nt (12). Vector sequences were trimmed from SAGE sequence with cross_ match (12). Sequence information in the processed chromatograms was converted into text files, and the SAGE tags were identified, extracted, and quantified using SAGE 2000 software, version 4.12 (http://www.invitrogen.com/sage). Tag nucleotide sequences and frequency data were output to Microsoft Access database tables for analyses. To determine differential expression, tag frequencies between both SAGE libraries were analyzed for significant distribution differences (P < 0.05) using statistical tools in the SAGE 2000 software based on a chi-square analysis combined with Monte-Carlo simulations (4).
Annotation of SAGE tags.
For annotation, the NlaIII recognition sequence "CATG," located 5' of each tag sequence, was appended to the 5' end of each SAGE tag to yield a 14-bp tag for use in database comparisons (40). Initially, cross-species comparisons were performed with the human and mouse UniGene SAGEmap databases (ftp://ftp.ncbi.nih.gov/pub/sage/map) (21). In addition, BLAST comparisons of SAGE tags to the National Institute for Biotechnology Information (NCBI) Nucleotide Sequence nonredundant (GenBank nt) database and the species-specific porcine gene index (SsGI, version 6.0) available through The Institute of Genome Research (TIGR, Rockville, MD; http://www.tigr.org/tdb/tgi/ssgi) databases were performed (2, 31). A perfect 14-bp match was required for inclusion in the final data set.
Confirmation of tentative consensus sequence for STAR.
Reverse transcription of 1 µg total RNA from filamentous conceptuses was performed at 42°C for 50 min in a 20-µl reaction containing: 500 ng oligo (dT)1218 (Invitrogen), 0.5 mM dNTPs, 1x first-strand buffer (Invitrogen), 0.01 M dithiothreitol (DTT, Invitrogen), 40 U RNasin ribonuclease inhibitor (Promega, Madison, WI), and 200 U SuperScript II reverse transcriptase. Semi-nested PCR primers for STAR were designed accordingly, two forward primers F1 and F2 and one reverse primer R2 (Supplemental Table S1, available at the Physiological Genomics web site).1
The initial PCR was carried out with 2 µl of reverse transcription product and 0.1 µM F1 and R2 primers in a 100-µl reaction volume containing 0.1 mM dNTPs, 1x PCR buffer (Eppendorf, Westbury, NJ), and 1.5 U HotMaster Taq DNA polymerase (Eppendorf). The semi-nested reaction mixture contained the same components as the initial PCR except 1 µl of the F1/R2 PCR product was used as template and the F2 primer was substituted for F1. The PCR conditions for the initial and semi-nested were as follows: denaturation at 94°C for 2 min followed by amplification (94°C for 30 s, 60°C for 30 s, 70°C for 3 min) for 32 or 16 cycles, respectively. The PCR product was sequenced as described above.
Real-time RT-PCR.
The template for real-time RT-PCR consisted of conceptus total RNA obtained from the same RNA pools used for SAGE libraries construction along with two additional independent total RNA preparations derived from pools of nonrelated ovoid and filamentous conceptuses collected on separate days. Porcine-specific sequences for CYP11A1, CYP19A, CYP17A1, STAR, HSD171B, MGST1, SOD1, manganese superoxide dismutase (SOD2), catalase (CAT), glutathione peroxidase 1 (GPX1), and ß-actin (ACTB) were obtained through GenBank (Supplemental Table S1). A two-step real-time RT-PCR method was employed, and real-time PCR was performed with the ABI Prism 7000 sequence detection system (Applied Biosystems). Reverse transcription was performed with 1 µg total RNA prepared from three individual ovoid or filamentous conceptus pools using the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA). The reverse transcription product was diluted 1:7, and 4 µl was used for PCR amplification with the iTaq SYBR Green Supermix with ROX kit (Bio-Rad). Each sample was assayed in triplicate according to the manufacturers protocol, except for a reduction of the total reaction volume to 25 µl. PCR conditions were as follows: denaturation 95°C for 10 min followed by amplification (95°C for 15 s and 60°C for 1 min) for 30 cycles. Preliminary data demonstrated that the ACTB transcript was constitutively expressed in ovoid and filamentous conceptuses. Therefore, ACTB was used as an endogenous control. To ensure amplification of a single product, gel analysis and a dissociation curve PCR were also performed. The amplification efficiency [10(1/slope) 1] of all the primers was similar. The relative quantity of the transcript was determined from the threshold cycle (CT) values using the Relative Quantification (RQ) software, version 1.1 (Applied Biosystems). To obtain the RQ value, the CT [difference between the endogenous control (ACTB) CT and target transcript CT] and
CT [difference between the
CT of a sample selected as the calibrator (a D11 sample) and the
CT of each individual sample] were calculated. The
CT of the calibrator was 0.0; the
CT of the other samples were increased or decreased relative to the calibrator. The RQ value was obtained by the following formula 2
CT. Analysis of RQ value for statistical significance was performed via a one-way ANOVA using the general linear model [Statistical Analysis System (SAS) software, version 8.02; SAS Institute, Cary, NC].
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The relative tag frequencies obtained via SAGE reflected the expression level of the transcripts represented by those tags. Examining the rate of occurrence of unique tags revealed that 0.8% (118/14,464) and 0.9% (124/13,098) of the transcripts had an expression level of
0.1% (i.e., 1 tag per 1,000 sequenced), in the ovoid and filamentous libraries, respectively. Comparing the unique SAGE tag frequencies between the combined libraries indicated that
1.9% (431/22,494) of total unique putative transcripts were expressed at significantly different levels (P < 0.05), with 233 tags (1.0% of 22,494) upregulated and 198 tags (0.9% of 22,494) downregulated in the filamentous conceptus. At a P < 0.001 level of significance, 87 tags representing unique genes were differentially expressed with 56 upregulated and 31 downregulated in the filamentous conceptus.
The current NCBI SAGE data repository, SAGEmap, does not contain porcine information; therefore, a cross-species comparison between the porcine SAGE tags and the human or mouse SAGEmap databases was initially evaluated. Of the 22,494 unique porcine conceptus tags, 9,024 and 13,207 matched a UniGene cluster in the human and mouse SAGEmap databases, respectively. A total of 6,409 tags matched the same UniGene cluster in the human and mouse SAGEmap databases. Considering the 431 differentially expressed pig conceptus tags (P < 0.05), 170 (39%) matched a tag sequence annotation in the human SAGEmap database.
The reliability of the differentially expressed tag annotations assigned by the human SAGEmap database was analyzed by a comparison with the SAGE tag annotation obtained from the GenBank nt and TIGR databases. BLAST sequence alignments of the 431 differentially expressed tags to the porcine gene index yielded matches to 278 (65%) tentative consensus (TC) sequences, albeit, some tags matched more than one TC. However, only 32 tags were assigned the same gene following the comparison of SAGEmap and TIGR tag annotations; the NCBI database search yielded results comparable to TIGR. Consequently, SAGEmap annotations were discarded due to greater consistency of tag annotation derived from GenBank nt and the TIGR porcine gene index. The genes reported herein were restricted to tags with only one gene assignment. In the end, 153 of the original 431 (35%) differentially expressed porcine SAGE tags were assigned a gene annotation corresponding to the human ortholog assignment as proscribed by LocusLink (29).
Tags were classified as being abundant if they constituted 0.5% of the total number of tags. Fifteen tags met the criteria, of which seven annotated to a single tentative consensus in the porcine gene index (Table 1). Five tags, annotated as stratifin (SFN), interleukin-1ß (ILB1), HSD17B1, cytochrome B oxidase subunit II (MTCO2L), and ribosomal protein L39 (MRPL39), were more abundant in filamentous conceptuses, whereas tags for cytokeratin 8 (KRT8) and cytokeratin 18 (KRT18) were more abundant in ovoid conceptuses.
|
|
|
The tag frequency comparison between ovoid and filamentous SAGE libraries indicated that CYP11A1, CYP17A1, CYP19A, HSD17B1, STAR, MGST, and SOD1 were more abundant in the filamentous (Table 1 and 2). Because of the presence of multiple transcript isoforms of CYP19A and STAR, primers capable of detecting all variants were used for real-time PCR (10, 28). The real-time PCR relative quantification (RQ) data is presented in Table 3. Statistical analysis of the RQ values confirmed the upregulation of CYP19A, STAR, CYP11A1, MGST1, GPX1, and SOD1 transcripts in the filamentous conceptus (Table 3). In contrast, SOD2 was downregulated in the filamentous conceptus compared with ovoid conceptus (Table 3). Differential expression of CAT, CYP17A1, or HSD17B1 was not detected by real-time PCR.
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
To understand the biology of pig conceptus development, it will be important to consider the functions ascribed to the genes identified in the present study and to determine their potential physiological role(s) during the elongation process. For example, the secretion of estrogens by the pig conceptus and their requirement for the maternal recognition of pregnancy has been well established (25, 26, 30). Previous work has demonstrated that increased embryonic estrogen secretion is accompanied by an elevation of CYP17A1 and CYP19A transcripts, encoding enzymes involved in the metabolism of cholesterol (43). The present results extend these prior studies by revealing the presence and upregulation of additional steroidogenic enzyme gene transcripts, CYP11A, HSD17B1, and the major regulator of steroidogenesis in nonconceptus tissue, STAR (36). Analysis of the SAGE tag frequencies within the ovoid and filamentous libraries revealed that HSD17B1 was the most abundant steroidogenic enzyme gene transcript in conceptuses of both stages. A transcript for 3ß-hydroxysteroid dehydrogenase (HSD3B1) was not detected in the SAGE tag population analyzed; however, RT-PCR with gene-specific primers demonstrated that the HSD3B1 transcript is present in the conceptus but at very low levels (data not shown). It is possible that HSD3B1 is represented in the population of differential expressed SAGE tags not analyzed herein (i.e., those present <10 times in either library).
The detection of STAR and its differential expression between ovoid and filamentous conceptuses is particularly significant because STAR is the penultimate mediator of steroid synthesis in nonconceptus tissue, yet its mRNA expression had not been previously reported in the pig conceptus (36). The STAR protein mediates the intramitochondrial transport of cholesterol (36). It is noteworthy that transcripts encoding all three rate-limiting proteins of estrogen synthesis, STAR, CYP11A1, and CYP19A, were detected in both ovoid and filamentous conceptuses, but were all more abundantly expressed in the elongated filamentous stage. If the resulting enzymatic activities of these steroidogenic proteins reflect their expression at the mRNA level, then increasing estrogen production in the filamentous conceptus could occur through a combination of the following mechanisms: 1) increased intramitochondrial cholesterol transport; 2) increased cholesterol metabolism; and 3) increased conversion of androgens to estrogen.
SAGE also revealed the differential expression of several genes involved in managing ROS during conceptus elongation (see Tables 2 and 3). Under normal circumstances, ROS are byproducts of metabolic processes active within the conceptus and its surrounding environment (7). Increased rates of development obtained from culturing embryos in the presence of ROS scavengers under low oxygen tension suggest that oxidative stress must be regulated for optimal development to occur (38). The potential significance of the differential expression of the antioxidant enzyme transcripts in the elongating pig conceptus is unclear. Interestingly, SOD1 and SOD2 are regulated oppositely in the filamentous conceptuses; SOD1 (cytoplasmic form) is upregulated, whereas SOD2 (mitochondrial form) is downregulated. Both forms of SOD convert the superoxide oxygen anion (O2) to H2O2, a less reactive oxidant, and under certain conditions SOD2 can compensate for the loss of SOD1 activity (13, 14). The other antioxidant enzymes detected, CAT, GPX1, and MGST1, are involved downstream of the SODs in the metabolism of H2O2 and modulating glutathione homeostasis.
It is tempting to speculate that interplay between H2O2 and MGST1 may elicit signal transduction responses that affect cytoskeleton changes within the conceptus during its dramatic elongation between developmental days 11 and 12. Although high levels of H2O2 are often cytotoxic, low concentrations of H2O2 have been shown to regulate normal physiological processes (3, 33, 44). For example, H2O2, may be involved in signal transduction pathways that induce changes in cell shape via the reorganization actin and promote proliferation as well as angiogenesis (3, 25, 44). The balance between ROS-stressed and nonstressed isoforms of glutathione S-transferase regulate stress kinases and protect the cell against H2O2-induced death (9, 44). Therefore, it will be important to further define the potential role(s) of ROS metabolism in mediating conceptus developmental potential and viability during pig elongation and implantation.
Regulating cellular morphology and proliferation during conceptus development and growth is also dependent on the cells ability to advance through several distinct cell cycles. Elongation of the pig conceptus initially involves cellular differentiation and remodeling prior to cellular proliferation (15, 30). SAGE revealed a group of three highly expressed genes, SFN, KRT8, and KRT18, that could potentially mediate cellular proliferation during pig conceptus development. The protooncogene p53A mediates G1 cell cycle arrest by stimulating SFN gene transcription (20). The first keratins detected during mouse embryogenesis are KRT8 and KRT18, and by the blastocyst stage these two proteins are associated with the differentiating trophoblastic epithelium (17). It is noteworthy that KRT18 also interacts with SFN to facilitate redistribution of SFN from the cytoplasm to the nucleus, a phenomenon that has been as associated with mitotic arrest (19). Therefore, in addition to its structural role within the cytoskeleton, KRT18 may also play a role in signal transduction via its association with SFN that may affect cell proliferation during elongation of the porcine conceptus (8, 19).
The present SAGE analyses established comprehensive gene expression profiles in ovoid and filamentous pig conceptuses undergoing normal development and elongation. The capacity to simultaneously measure gene expression within an entire metabolic pathway, demonstrated by our downstream analysis of the estrogen synthesis and ROS management pathways, underscores the potential power and validity of SAGE for investigating conceptus development in swine. Ongoing work in our laboratory is elucidating the transcriptomes of pig conceptuses prior to and during intermediate, tubular, stages of elongation in an attempt to more fully understand temporal gene expression events and their interactions during the elongation process. Furthermore, it will be important to compare these gene expression profiles obtained from normal conceptuses with those representing aberrant or in vitro produced conceptuses to more fully correlate specific transcriptional events with efficient conceptus development. Ultimately, such knowledge will be useful for the establishment of comprehensive physiological models of conceptus development and provide the basis to devise new strategies aimed at enhancing reproductive efficiency in swine.
![]() |
GRANTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
ACKNOWLEDGMENTS |
---|
Present address of E. L. Long: National Cancer Institute, Center for Cancer Research, Mammary Biology and Tumorigenesis Laboratory, Bldg 10, Rm 5847, Bethesda, MD 20892-1402.
![]() |
FOOTNOTES |
---|
Address for reprint requests and other correspondence: L. A. Blomberg, Biotechnology and Germplasm Laboratory, USDA Agricultural Research Service, Animal and Natural Resources Institute, Bldg 200, Rm 101A, BARC-East, Beltsville, MD 20705 (E-mail: lblomberg{at}anri.barc.usda.gov).
10.1152/physiolgenomics.00157.2004.
1 The Supplemental Material for this article (Supplemental Tables S1 and S2) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00157.2004/DC1.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Visit Other APS Journals Online |