| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Animal Reproduction and Biotechnology Laboratory, Colorado State University, Fort Collins, Colorado 80523
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. One novel cDNA fragment did not share sequence identity to previously reported genes, except for a similar DNA sequence in the human genome. AIF-1 mRNA was present in developing placenta through Gestation Day 36, and abundant levels were observed in adult bovine spleen and lung. The novel gene, which we have named periattachment factor (PAF), was not detected in adult tissues using Northern analysis or in conceptuses between Days 30 and 36 of pregnancy. Additional sequence information for bPAF was obtained from a cDNA library constructed from a 25-day bovine embryo. The protein corresponding to the open reading frame has four protein kinase C phosphorylation sites, two casein kinase II phosphorylation sites, a nuclear targeting sequence, but no obvious DNA or RNA binding motifs. Abundant expression of this gene during a narrow but critical window of embryonic development makes it worthy of further study.
developmental biology, early development, embryo, pregnancy, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, which mediates maternal recognition of pregnancy [10]; and 3) placental lactogens, which are secreted from fetal binucleate cells after the bovine embryo has attached to the endometrium [11–13]. Studying developmentally regulated gene expression in preimplantation and peri-implantation embryos has been problematic because of a paucity of tissue. To better understand changes in gene expression in bovine embryos between Gestation Day 15.5 and 17.5, we used the reverse transcription-polymerase chain reaction (RT-PCR)-based technique developed by Liang and Pardee [14], mRNA differential display. This technique has been used by others to study earlier stages of bovine preimplantation embryos in various contexts [15–17] but not in elongating stage embryos. This technique not only allows detection of changes in expression of known factors present in peri-implantation embryos but also unknown potential mediators of significant events that occur during this period of development, such as maternal recognition of pregnancy. Further experiments were conducted to obtain sequence information and determine patterns of expression in bovine embryos and bovine adult tissues for the cDNA fragments that were differentially expressed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All procedures with animals were done using protocols approved by our Institutional Animal Care and Use Committee. Cows were inseminated artificially 12 h post estrus (Day 0 = estrus), and conceptuses were collected on Pregnancy Days 15.5 and 17.5. Cows were confined in a cattle chute and given an epidural block of 2% procaine. A sterile 24-gauge Foley catheter was inserted through the cervix into the uterine body, and modified Dulbecco phosphate-buffered saline supplemented with 0.1% BSA was flushed through the uterus nonsurgically [18]. Upon recovery, 15.5-day embryos were measured, rinsed with sterile modified Dulbecco phosphate-buffered saline plus 0.1% BSA, and frozen in liquid nitrogen. Four 15.5-day embryos were recovered with lengths of 1.0, 3.8, 4.0, and 4.2 cm. We chose to use the 3.8- and 4.0-cm embryos for differential display. Two 17.5-day embryos were recovered similarly. These exceeded 10 cm in length; we did not attempt to measure their precise lengths because stretching and tearing made precise measurement impossible. All embryos were from nonsuperovulated cows except for one set of 15-day embryos used for independent verification of differential expression as described later.
Messenger RNA Differential Display Technique
Total RNA Isolation Rather than using pools of embryos, we chose to analyze individual embryos and selected bands to study that were differentially expressed consistently. Total RNA from two 17.5-day and two 15.5-day embryos, 3.8 and 4.0 cm in length, was isolated [19] with the following modifications. Lysis buffer with the addition of 25 µg poly(C) (Boehringer Mannheim, Indianapolis, IN) was added to each embryo, 1 ml for 17.5-day embryos and 0.5 ml for 15.5-day embryos. Embryos were sheared through a 22-g needle, and RNA was extracted [19]. To remove contaminating DNA, samples were treated for 30 min at 37°C with 3 U RQ1 RNase-Free DNase (Promega, Madison, WI) in the recommended buffer. Following DNase treatment, 200 µl of diethyl pyrocarbonate-treated water were added to each sample, and RNA was extracted in phenol-chloroform-isoamyl alcohol [19]. The aqueous phase was precipitated, resuspended, and the concentration of RNA calculated by determining the A260. RNA was diluted to 0.25 µg/µl, aliquoted (0.5 µg) into 0.5-ml microcentrifuge tubes and stored at -70°C.
Reverse Transcription Total RNA was reverse transcribed using 1 of 12 5'-T11MN-3' primers (where M represents A, C, or G, and N represents all four nucleotides); each oligo (dT11)+2 base pair primer was used in a separate reaction. Total RNA (0.5 µg) and oligo (dT11) primer (20 µM) were combined, and the volume was brought to 12.4 µl with diethyl pyrocarbonate-treated water. Samples were incubated for 10 min at 70°C and cooled to 45°C. The remaining components were combined in a mixture, prewarmed to 45°C, and added to the samples yielding a final volume of 20 µl and final amounts of 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 2.5 mM MgCl2, 0.1 mg/ml BSA, 10 mM dithiothreitol, 40 U RNAsin (Promega), 0.2 mM of each deoxynucleotide triphosphate (dNTP), and 200 U Superscript RT II enzyme (Gibco BRL, Gaithersburg, MD). Samples were incubated at 45°C for 1 h, 95°C for 5 min, and cooled to 4°C.
PCR Condensate was collected by centrifugation at 16 000 x g for 30 sec. For each 20-µl PCR mixture, 2 µl of each RT reaction were added to 18 µl of the mixture described above to obtain the final amount of 1x Vent polymerase buffer (New England Biolabs, Beverly, MA), 2 µM of each dNTP, 5 µCi of [35S]-deoxy-ATP (1000 Ci/mmol; Amersham, Cleveland, OH), 0.5 U Vent exo-polymerase (New England Biolabs, Beverly, MA), and 0.5 µM of 1 arbitrary decamer. Five unique arbitrary decamers were used in separate PCR reactions for each of the 12 reverse-transcribed preparations. Samples were overlaid with 40 µl mineral oil and subjected to 40 cycles in a model 9600 thermocycler (Perkin-Elmer/Cetus, Norwalk, CT) using the following parameters: 94°C for 30 sec, 42°C for 60 sec, and 72°C for 30 sec. Radiolabeled PCR product (3 µl) was combined with 3 µl stop solution (USB Sequenase, version 2.0, T7 DNA Polymerase, Cleveland, OH), boiled for 2 min to denature the DNA, placed on ice, and separated on a 6% polyacrylamide sequencing gel. The gel was exposed to Kodak X-Omat film (Eastman Kodak, Rochester, NY) for 12 h, and fragments that were consistently differentially expressed (Fig. 1) between the two embryonic stages were identified. If differentially expressed fragments were present, the gel was exposed to Kodak X-Omat film for an additional 12 h. A razor blade was used to remove the bands present in greater abundance in 17.5-day, compared with 15.5-day embryos from the film. Once removed, the x-ray was realigned with the acrylamide gel, and DNA fragments were excised with a razor blade using the x-ray film as a template. Fragments were placed in 100 µl sterile water in 0.5 ml microcentrifuge tubes, and the acrylamide gel was re-exposed to Kodak X-Omat film for 12 h to guarantee that the correct band was excised. Tubes containing the cDNA fragments were boiled for 15 min and centrifuged for 5 min at 16 000 x g to pellet debris. The supernatant was transferred to a 0.5-ml microcentrifuge tube; 8 µl of 3 M potassium acetate, 2.5 µl of glycogen (20 mg/ml), and 300 µl of 100% isopropyl alcohol were added, and the DNA was precipitated, collected by centrifugation, and resuspended in 10 µl water.
|
Reamplification of PCR Products To reamplify the differentially expressed product, DNA (5 µl) was added to 95 µl of PCR solution to obtain a final amount of 2.5 µM of the respective oligo (dT) primer, 1 x Vent polymerase buffer, 20 µM of each dNTP, 1.25 U Vent exopolymerase, and 2.5 µM of the respective decamer. Samples were overlaid with mineral oil, and the same PCR parameters were used as described previously except the last cycle was followed by a 72°C extension time for 10 min.
Southern Analysis
Southern analysis was used to identify cDNA fragments that were unique and are not represented in multiple differential display analyses. After reamplification, DNA from each PCR reaction (50 µl) was separated on a 2% agarose gel, denatured in 0.5 M NaOH/1.5 M NaCl for 30 min, neutralized in 0.5 M Tris (pH 7.5)/1.5 M NaCl for 30 min, and transferred overnight by capillary action to nylon membrane (Schleicher and Schuell, Keene, NH). The membrane was cross-linked at 1200 J using a CL-1000 Ultraviolet Crosslinker (UVP, Upland, CA) and prehybridized for 2 h at 42°C in 10 ml of solution containing 50% formamide, 5 x Denhardt solution, 5 x saline sodium citrate (SSC), 0.5% SDS, 5 µg poly (A) RNA, and 1.5 mg boiled salmon sperm DNA. DNA (25 ng) from each PCR reaction was radiolabeled with [32P]-deoxy-cytidine 5'-triphosphate (dCTP) (3000 Ci/mmol; ICN, Costa Mesa, CA) using Random Primers DNA Labeling System (Gibco/BRL). Following incubation, 1 x 107 cpm of probe was denatured in 0.3 M NaOH for 5 min, neutralized with 0.05 M Tris (pH 7.2)/0.3 M HCl, and immediately added to the Southern blot. After hybridizing overnight at 42°C, the membrane was rinsed for 20 min at 42°C in 2 x SSC/0.1% SDS, 15 min at 65°C in 1 x SSC/0.1% SDS, and 10 min at 65°C in 0.1 x SSC/0.1% SDS to remove nonspecifically bound probe. Then the membrane was exposed to Kodak X-Omat film for approximately 1 h. If a single band was present on the Southern, the corresponding cDNA fragment was used to probe the following Northern blots. If multiple bands were present, the largest cDNA fragment was identified and used in the following Northern analyses.
Verification of Differential Expression Between 15.5-Day and 17.5-Day Bovine Embryos by Northern Blots
Embryo Collection Northern analysis was conducted with a new set of embryos to verify completely independently, an increase in gene expression between Days 15.5 and 17.5 of pregnancy as detected by differential display. Because of the relatively small amount of tissue present on Day 15.5 of pregnancy, six cows were superovulated [20] with FSH-P (Schering, Kenilworth, NJ), and 18 embryos were collected [18] on Day 15 of pregnancy, ranging from 0.2 to 6.3 cm in length. However, we used only the 12 embryos that were >2 cm in length for the pool to match the sizes of the embryos from nonsuperovulated cows used for the differential display.
Embryos were collected on Day 17.5 of pregnancy from two cows that were not superovulated; these embryos were used individually. All embryos collected were rinsed and frozen as described in the differential display technique section above.
Northern Blot Analysis Total RNA from the pool of 12 additional 15-day embryos ranging in size from 2 to 6 cm and two additional 17.5-day embryos was isolated [19], resuspended in loading buffer, and heated for 5 min at 70°C. RNA was separated on a 0.66 M formaldehyde/1% agarose denaturing gel. Gels were rinsed 4 x for 5 min in water and transferred by capillary action to nylon membranes overnight. Membranes were cross-linked at 1200 J as described earlier for the Southern procedures.
Differentially expressed fragments chosen from the Southern blots were radiolabeled with (32P)-dCTP (3000 Ci/mmol, ICN) as described for Southern blots, and used to probe Northern blots. Prehybridization and hybridization of probe to membranes were described in the above Southern blot section. After hybridization, membranes were rinsed for 30 min at 42°C in 2 x SSC/0.1% SDS, 30 min at 50°C in 0.5 x SSC/0.1% SDS, and 15 min at 50°C in 0.1 x SSC/0.1% SDS. Membranes were exposed to Kodak X-Omat film for 24–72 h and exposed to a PhosphorImaging screen (Molecular Dynamics, Sunnyvale, CA) for 8 and 16 h. Resulting images were analyzed after subtracting background values using the Molecular Dynamics software. The results were used to quantify differences in steady state mRNA content between Day-15 and Day-17.5 bovine embryos.
Complementary DNA fragments encoding mRNAs demonstrated to be differentially expressed between embryonic stages were further cloned and sequenced. After probing with the differentially expressed fragments, membranes were stripped by soaking in boiling 0.1 x SSC/0.1% SDS for 20 min. The metabolic gene, IB15 cyclophilin [21], was radiolabeled and hybridized to the membranes to normalize for loading differences; bands were quantified by PhosphorImaging.
Cloning and Sequencing
Four differentially displayed fragments identified from the previous experiments were reamplified using their respective primers and cloned using a TA cloning kit (Invitrogen Corp., Carlsbad, CA) according to the manufacturer's instructions. A modified minialkaline-lysis/PEG precipitation procedure (Perkin-Elmer) was used to prepare plasmid DNA for sequencing. The Macromolecular Resources Service at Colorado State University sequenced plasmid DNA containing each differentially expressed fragment. Fragments were restriction digested to manageable sizes for sequencing, including appropriate overlaps, and sequenced in both directions. The sequences were compared with known mammalian sequences using the BLAST algorithm [22]. From the Northern blot and sequence data, two differentially displayed fragments, allograft inflammatory factor-1 (AIF-1) and fragment 7, were used to complete the following experiments. Fragment 7 has been given the name bovine periattachment factor (bPAF).
Embryonic Gene Expression During Early Pregnancy
Embryo Collection Bovine embryos were collected on Days 17.5, 30, 33, 35, and 36 of pregnancy. Day 17.5 embryos were collected, rinsed, and frozen according to procedures described previously. Later-stage embryos were collected according to the following nonsurgical procedure. A metal tube 40 cm in length with an outer diameter of 7 mm and an inner diameter of 6 mm and a stainless steel cervical expander measuring 48.5 cm in length with a diameter of 6 mm tapering to 3 mm were buffed to remove all rough edges. After sterilization, the expander was placed inside the metal rod so that the expander tip was exposed. Pregnant cows were given an epidural block of 5 ml of 2% procaine, and the expander and rod were gently threaded through the cervix; once inside the internal os, the expander was slowly removed leaving the hollow rod in place. Tygon tubing S-50-HL (Fisher Scientific, Pittsburgh, PA) (60 cm) with an inner diameter of 0.63 cm was placed over the metal tube, and a 60 ml syringe filled with modified Dulbecco PBS plus 0.4% BSA was anchored to the opposite end of the Tygon tubing. The entire contents in the syringe were expelled into the uterus, and the uterus was gently massaged to dislodge the embryo. While applying gentle pressure, uterine contents were aspirated into the syringe. This procedure was repeated until the embryonic tissue was recovered. Conceptus tissues were rinsed, and extraembryonic membranes were frozen as described previously; the embryo proper was discarded.
Northern Blots for Embryos Obtained During Early Pregnancy After isolating RNA from individual embryos, total RNA (20 µg) was separated by gel electrophoresis and capillary transferred to nylon membranes. The partial cDNA for AIF-1 was radiolabeled as described previously and used to probe the membrane. After rinsing away excess probe, the membrane was exposed to Kodak X-Omat film for 72 h, and further exposed to PhosphorImaging analysis for 24 and 48 h to quantify the level of expression among embryos. After collecting data for bAIF-1, the membrane was stripped, and this procedure was repeated using radiolabeled bPAF cDNA to probe the membrane. The membrane was exposed to Kodak X-Omat film for 96 h and analyzed by densitometry, and/or exposed for PhosphorImaging analysis.
To provide an internal control, membranes were stripped and hybridized to a radiolabeled cDNA fragment encoding the 28S ribosomal RNA (Ambion, Austin, TX) [23]. Membranes then were rinsed to remove excess probe, exposed to Kodak X-Omat film for 20 min and further exposed to PhosphorImaging analysis for 5 and 15 min to quantify the steady state content of 28S ribosomal RNA.
To verify information obtained using the 28S ribosomal subunit as an internal control, membranes were stripped and hybridized to radiolabeled cDNA encoding human 
-actin [24]. Membranes were rinsed, exposed to Kodak X-Omat film for 6 h and further exposed to PhosphorImaging analysis for 30 min and 1.5 h to quantify the steady state content of -actin mRNA.
Northern Analyses to Determine Differences in Gene Expression Among Various Adult Tissues
Tissue Collection, RNA Isolation, and Northern Blot Procedures
Adult tissues including brain, heart, kidney, liver, lung, anterior pituitary, skeletal muscle, and spleen were collected immediately after slaughtering a heifer, dissected into pieces (
1 g) and frozen in liquid nitrogen. Total RNA was isolated, and Northern blots were performed as described above.
Cloning AIF and bPAF cDNA from a 25-Day Bovine Embryo cDNA Library
The 3' partial bPAF and bAIF cDNA fragments were used to screen a bacteriophage lambda, Uni-ZAP cDNA library (Stratagene, La Jolla, CA) prepared from 25-day bovine embryos [25]. Approximately 800 000 plaques were screened according to the manufacturer's instructions. Positive clones were isolated, and pBluescript phagemids were excised by the in vivo excision protocol using ExAssist helper phage and the Escherichia coli SOLR strain (Stratagene). Rescued pBluescript SK-plasmids were restriction digested with Xho 1 and EcoR1, separated by gel electrophoresis, and hybridized to the original probe to verify their identity. Plasmids were prepared and sequenced by Macromolecular Resources using T3 and T7 primers and six specific primers synthesized for bPAF cDNA. Nucleotide and amino acid sequence searches were performed using the BLAST algorithm, and motif searches were generated using the Prosite and PSORT II.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RNA from two 15.5-day and two 17.5-day bovine embryos was reverse transcribed and amplified using 60 primer combinations resulting in 240 PCR reactions. Complementary DNA profiles generated using the same primer set in each embryo were compared side by side. Fifteen amplicons (designated bands 1–15) putatively present in larger amounts in 17.5-day than 15.5-day embryos by inspection, were excised, reamplified, and used as radiolabeled probes in subsequent Southern and Northern analyses (Table 1). Figure 1 illustrates differential displays of four amplicons obtained with separate primers that are present in greater abundance in 17.5-day embryos (lanes 3 and 4) compared with 15.5-day embryos (lanes 1 and 2).
|
Differentially expressed amplicons were reamplified by PCR using the appropriate primer set. The resulting reactions were used in two ways: 1) amplified cDNA fragments were separated on a 1% agarose gel and transferred to nylon membrane; 2) the same cDNA fragments were radiolabeled and used as molecular probes to identify which amplicons were unique on the Southern blot. Several cDNA fragments appeared to be identical, even though they were PCR amplified using different primer sets (Table 1).
Complementary DNA fragments represented by a single band on the Southern blot (bands 1, 2, 3, 4, 5, and 13) were chosen and used in the following Northern blot analyses. In samples for which several cDNA fragments appeared to represent the same upregulated gene, the largest amplicon was chosen and used in subsequent Northern blot analyses (bands 6 and 7).
Northern Blot Analysis With Differentially Displayed cDNA Fragments and Nucleotide Sequencing Analysis
Northern analyses were conducted to confirm that the mRNA concentration of differentially displayed cDNA fragments was greater in 17.5-day than 15.5-day bovine embryos. Total RNA from a pool of 15-day embryos and two 17.5-day single embryos was screened using 32P-labeled cDNA fragments (bands 1, 2, 3, 4, 5, 6, 7, and 13) from PCR mixes. Of the eight cDNA fragments that appeared to be differentially expressed, four (bands 1, 2, 6, and 7) clearly had a higher mRNA concentration in 17.5-day than 15.5-day embryos (Table 1); equal concentration was noted for three fragments, and no signal was detected for one cDNA fragment.
To ensure that PCR-generated fragments were not yielding misleading results, the four differentially expressed cDNA fragments were subcloned into the plasmid pCRII vector; inserts were excised, radiolabeled using (32P)-dCTP, and used to reprobe the Northern blots containing total RNA isolated from both stages of embryos. In all cases, equivalent results were obtained from Northern blots probed with fragments generated by PCR or subcloned inserts.
A single band 800 bp in length was detected in all samples (Fig. 2) when using fragment 1 as a probe. Expression of this transcript was 4.3-fold and 4.9-fold greater in individual 17.5-day embryos than the 15-day pool. Three distinct mRNA transcripts were noted for fragment 2, a major mRNA species 5000 bp in length, and two minor transcripts 4500 bp and 2000 bp in length (Fig. 2). The major transcript was present on the Northern blot in 11.3-fold and 10.6-fold more abundance in the two 17.5-day than 15-day embryos; the 4500 bp transcript was 4.2-fold and 9-fold more abundant in 17.5-day embryos, compared with the 15-day pool. Intensity of the smallest band (2000 bp) was not measured because of a break in the agarose gel that separated the hybridization band. Fragment 6 yielded a single mRNA transcript 1600 bp in length, and relative content was 2.5-fold more in 17.5-day than 15-day embryos (Fig. 2). Finally, a single transcript of 1800 bp in length was shown for cDNA fragment 7; it was 9.4-fold and 11.7-fold more abundant in the two 17.5-day embryos than the 15-day embryo pool (Fig. 2). RNA loading was normalized among all samples using cyclophilin cDNA.
|
The four subcloned differentially expressed cDNA fragments were sequenced in both directions, and results were compared with known mammalian sequences using BLAST. Three cDNA fragments had significant sequence identity to known genes (bands 1, 2, and 6), and one cDNA fragment demonstrated no sequence homology (band 7) to previously reported genes except for a similar sequence found in the human genome (XM-166586). The nucleotide sequence for cDNA fragment 1 was 87% identical to bases 175–422 of human AIF-1 (GenBank accession no. U19713), and 84% identical to bases 236–483 of rat AIF-1 (GenBank accession no. U17919). Complementary DNA fragment 2 had 79% sequence identity to bases 2662–2853 of human ligands of eph-like receptor kinase (LERK)-5 mRNA (GenBank accession no. U81262) (Fig. 3). Complementary DNA fragment 6 had 97% sequence identity to the 3'-end of bovine interferon- (bIFN) (GenBank accession no. M31557), an abundant and well-characterized gene product responsible for maternal recognition of pregnancy. Two cDNA fragments (bands 2 and 6) were not pursued in the following experiments. Complementary DNA fragment 2 had large multiple mRNA transcripts that may increase the difficulty in obtaining additional nucleotide sequence information, and bIFN (band 6) has been well characterized in cattle. In the following experiments, we used cDNA fragment 1 (AIF-1) and cDNA fragment 7 (bPAF).
|
Northern Blot Analysis Identifying Continued Presence of cDNA Fragments Through Day 36 of Pregnancy
Single conceptuses were collected from cows on Days 17.5, 30, 33, 35, 36 of pregnancy to determine the presence of mRNA for bAIF-1 and bPAF in the developing placenta. These days of gestation were chosen because of ease of collecting embryonic tissues nonsurgically. Bovine AIF-1 was present as a single mRNA transcript of 800 bp in all samples as determined by Northern analysis using radiolabeled fragment 1 (Fig. 4). Expression of this gene, normalized to -actin mRNA, was 10-fold, 2.5-fold, 6.3-fold, and 7.6-fold greater in older embryos than mRNA levels present in the 17.5-day bovine embryo. Expression was 3.5-fold, 1.1-fold, 2.6-fold, and 2.4-fold greater than levels present in the 17.5-day bovine embryo when using 28S rRNA to normalize for loading differences. Likely -actin is more valid because large amounts of 28S rRNA present on the membrane may preclude binding of the 28S rRNA probe in a linear range. On the other hand, the -actin signal (Fig. 4c), but not the 28S rRNA signal, is indicative of overloading of 17.5-day RNA. Although these issues preclude precise quantification of bAIF-1 mRNA in these embryos, it clearly is present at all stages examined.
|
The bPAF cDNA also was used to probe the Northern blot (Fig. 4). A single band of 1800 bp was identified only in the RNA obtained from the 17.5-day embryo; no bands were identified in samples from 30-, 33-, 35-, and 36-day bovine embryos.
Northern Blot Analyses Identifying the Presence of Differentially Expressed Genes in Bovine Adult Tissues
Total RNA was isolated from bovine brain, heart, kidney, liver, lung, anterior pituitary, skeletal muscle, and spleen to determine whether the differentially expressed genes are present in adults. Northern blots were probed with 32P-radiolabeled bAIF-1 and bPAF cDNA; message levels were normalized to 28S ribosomal RNA. When probed with bAIF-1, a single mRNA transcript of 800 bp was identified in lung and spleen at levels 10-fold and 2.4-fold greater than in the 17.5-day bovine embryo; weakly hybridizing bands were also detectable in brain, kidney, and liver (Fig. 5).
|
A single mRNA transcript of 1800 bp in length was identified in the 17.5-day bovine embryo when the Northern blot was probed with 32P-radiolabeled bPAF cDNA. A very weakly hybridizating band also was present in kidney mRNA. No band was detected in remaining adult tissues (Fig. 5).
Complimentary DNA Library Screening and Nucleotide Sequence of Differentially Expressed bPAF cDNA
To determine the full-length nucleotide sequence of bPAF cDNA, a 25-day bovine embryo cDNA library was screened. Four independent clones were identified, indicating that this gene was still expressed on Day 25 of gestation; the nucleotide sequence for the longest clone (1673 bp insert) has been submitted to Genbank (accession no. AY027656), and the open reading frame is shown in Figure 6. Three shorter clones were identified from the 25-day bovine embryo library, p7 (718 bp insert), p1 (446 bp insert), and p10 (400 bp insert). The three shorter clones confirmed the sequence obtained from the longest clone. With the exception of a single mismatch, the 3' terminal regions of the clones were identical to bPAF cDNA obtained from the original differential display analysis. A possible polyadenylation signal existed 13–18 bases upstream from the poly(A) tail that contained a single base pair transition from an adenine to a thymidine residue (ATTAAA). The full-length sequence was compared with known nucleotide sequences using the BLAST computer program; no significant homology was found to previously published gene sequences other than a similar DNA sequence with an open reading frame in the human genome.
|
The full-length nucleotide sequence had a possible open reading frame of 381 bp from the possible ATG start site to a TAG stop codon, with a deduced protein sequence of 126 amino acid residues (Fig. 6). The cDNA clone contained a 625-bp 3' untranslated region. The molecular weight of the putative protein was calculated to be 13.5 kDa, and a search of protein databases using BLAST indicated that the deduced amino acid sequence was not homologous to previously reported amino acid sequences other than the theoretical one associated with the hypothetical human gene (XM-166586). From Prosite and PSORT II analyses, four putative protein kinase C phosphorylation sites and two casein kinase II phosphorylation sites as well as a nuclear targeting sequence as described in Fig. 6 were identified. The human homolog had 83% sequence identity and 77% amino acid sequence identity to bPAF.
Complimentary DNA Library Screening and Nucleotide Sequence of Differentially Expressed bAIF-1 cDNA
To determine the full-length nucleotide sequences of bAIF-1 cDNA, we screened the same 25-day bovine cDNA library in the same way as described in the previous section for bPAF. Four independent clones were obtained; the nucleotide sequence for the longest clone (686 bp insert) was submitted to GenBank (accession no. AF348450). With the exception of a 1-bp mismatch, the 3' terminal region of the clone was identical to the cDNA fragment used as a probe in this experiment and originally obtained from the differential display analysis. The full-length nucleotide sequence had a presumed open reading frame of 441 bp from the possible ATG start site to a TGA stop codon, with a deduced protein sequence of 147 amino acid residues. The cDNA clone also contained a 143-bp 3' untranslated region that included a typical polyadenylation signal [26] 17–22 bases upstream from the poly(A) tail. The full-length sequence was compared with known nucleotide sequences using the BLAST computer program; 88% and 84% sequence homology was identified to human and rat AIF-1 (GenBank accession nos. U19713 and U17919) at the nucleotide level. The deduced protein is 16.8 kDa; a search of Swiss-Prot and TREMBL databases using BLAST indicated that deduced amino acid sequence was 82% homologous to the amino acid sequence for human AIF-1 (GenBank accession no. P55008). Several potential amino acid motifs were determined using Prosite, including a single PKC phosphorylation site, five casein kinase II phosphorylation sites, two N-myristoylation sites, and one amidation site.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There were several advantages using mRNA differential display, compared with traditional subtraction library techniques for identifying mRNA species present in greater abundance on 17.5 days relative to 15.5 days of pregnancy. The mRNA differential display profiles could be compared, and differentially expressed cDNA fragments were easily identified, reamplified, subcloned, and sequenced; furthermore, cDNA fragments were used for Southern and Northern analyses. The mRNA differential display was extremely sensitive and required little tissue. Using this technique, we were able to detect a previously unidentified gene in cattle as well as genes that have been characterized but not identified during embryonic development.
The confirmation step using Northern blot analysis required more tissue than the mRNA differential display procedure. We therefore used superovulation to obtain a large number of 15-day embryos and selected those of similar size to the nonsuperovulated ones used. Embryos from superovulated cows result in normal calves [27] and pregnancy rates similar to those from nonsuperovulated donors [28]. Our study showed that both 15-day superovulated and nonsuperovulated embryos had lower mRNA content for the four genes studied than nonsuperovulated day-17.5 embryos, albeit, measured by different techniques.
There were some disadvantages using mRNA differential display, compared with traditional subtraction library techniques. Several bands were detected using different primers sets that represented the same gene, requiring additional Southern blot and Northern blot analyses to verify the presence of unique differentially expressed cDNA fragments (Table 1). Sometimes only the 3' untranslated region of the gene is recovered making sequence identification more difficult, and performing PCR reactions in duplicate using several different primer combinations is laborious and time consuming. Also, the abundance of bIFN mRNA during these stages of pregnancy may overwhelm the system, reducing the number of rare mRNAs detected. Despite these limitations, the procedure proved useful for the present analysis.
Complementary DNA fragment 2 (bLERK-5) was isolated from a single mRNA differential display that was amplified using a single primer set. Sequence information revealed that this cDNA fragment was PCR amplified from the oligo (dT) plus 2 bp primer, which anchored at both the 5' and 3' positions. Similarly, for bPAF, degenerate binding occurred between the oligo (dT) plus 2 bp primer and the template. In this case, the oligo (dT) plus 2 bp primer annealed to the novel cDNA at both the 5' and 3' positions; thus, the arbitrary decamers that were used did not always function effectively as primers. The results obtained for cDNA fragments encoding bLERK-5 and the novel gene product simply add another dimension to the anchoring scheme suggested by Liang and Pardee [14].
Bovine AIF-1
The nucleotide sequence for cDNA fragment 1 shared 88% and 84% sequence homology to human and rat AIF-1. AIF-1 originally was identified in rats that developed arteriosclerotic lesions during chronic rejection of cardiac allografts [29]. During the initial stages of cardiac rejection 30 to 75 days post transplantation, monocytes/macrophages expressing rAIF-1 accumulate in the vessels. Smooth muscle cells migrate into the intima in response to the secretion of many cytokines and growth factors present during the inflammation response. Finally, smooth muscle cells secrete extracellular matrix and soluble growth and chemotactic factors, resulting in dramatic cellular proliferation and vessel occlusion [30, 31].
A 4.5-fold increase in bAIF-1 mRNA content was demonstrated in 17.5-day relative to 15-day bovine embryos; bAIF-1 mRNA expression in extraembryonic membranes continued through Day 36 of pregnancy. Although a slight decrease in expression was seen on Day 33 of pregnancy, we believe that this was due to variation in expression from embryo to embryo because the other embryos were fairly consistent in this respect between Days 30 and 36. Similar to cardiac transplantation, an allograft forms between fetal and maternal tissues during implantation as bAIF-1 transcription is increasing; this may indicate an important role of bAIF-1 during pregnancy. At this time, uterine macrophages are abundant in rodents, women, and pigs [32, 33]. It has been suggested that there are several potential macrophage functions in reproduction including tissue remodeling events and a pivotal role in coordinating the local immune response to the fetus during pregnancy [33]. We cannot rule out that the detected bAIF-1 mRNA was from activated macrophages in the embryonic samples; however, we are unaware of evidence that macrophages are present in preimplantation and peri-implantation bovine embryos. We also do not know whether bovine uterine macrophages express bAIF-1 mRNA. If bAIF-1 mRNA detection is from activated macrophages in 15.5-day and 17.5-day bovine embryos, these macrophages either infiltrated into or bound to the embryonic trophoblast.
Preliminary evidence using in situ hybridization indicated that bAIF-1 mRNA was expressed from bovine trophoblast cells on 17.5 days of gestation (M. Glover, unpublished observations). Rat AIF-1 mRNA expression and antigen also are found in the differentiating germ cells of the rat testis, suggesting a distinct role in spermatogenesis not related to inflammation [29]; bAIF-1 also may have a role in embryonic cell differentiation.
Whether expression of bAIF-1 mRNA is from uterine macrophages or embryonic trophoblast, it may be stimulated by abundant levels of bIFN present during early pregnancy. Rat AIF-1 mRNA was upregulated 6-fold in vitro from rat macrophages after stimulation with interferon- (INF) [29]. IFN, a cytokine secreted by activated T lymphocytes and natural killer cells, is a potent activator of macrophages, affecting many macrophage properties and functions [34]. Although IFN and IFN are in separate classes of interferons and bind to different receptors, the downstream effects are often similar. Both types I and II interferons may control the regulation of bAIF-1 mRNA content.
Human AIF-1 mRNA content was the highest in human cells of lymphoid origin, in particular, human spleen, peripheral blood lymphocytes, and thymus, but lower in liver, lung, and placenta [35]. Abundant levels of bAIF-1 mRNA also were detected in bovine spleen. Unlike human tissue, the highest bAIF-1 mRNA content was in the lung, at least in this particular cow.
Bovine LERK-5
Fragment 2 was radiolabeled and used to probe Northern blots. Three distinct mRNA transcripts were observed; the major transcript was 5000 bp in length and present in 11-fold higher abundance in 17.5-day relative to 15-day embryos. This fragment shared 79% sequence identity to bases 2662–2853 of human LERK-5 mRNA (GenBank accession no. U81262).
Several ligands that exhibit extensive cross-binding to different members of the eph-like receptor tyrosine kinases (RTKs) have been identified [36, 37]; these have been termed LERKs, ligands of eph-like receptor kinases. RTKs have been grouped into families based on structural and functional similarities [38, 39]. The eph-like branch of RTKs is the largest family including more than 28 members present in several vertebrate species [39–42]. Although little is known about the function of specific RTKs, expression of eph-like RTKs during embryogenesis suggests a role in development [43, 44]. Also, overexpression of eph-like RTKs in tumor-derived cell lines implicate these receptors in oncogenesis [41, 45].
Complementary DNA clones have been isolated for human and murine LERK-5. Cerretti et al. [46] isolated a clone-encoding human LERK-5 that was 1777 bp in length with a potential open reading frame encoding a 333 amino acid peptide. Furthermore, a 2611-bp cDNA for murine LERK-5 was isolated from an 11.5-day mouse embryo library; it had an open reading frame of 336 amino acids that shared 97% sequence identity to human LERK-5 [46]. There is relatively little amino acid sequence identity among the various LERK proteins. LERK-5 peptide sequence is most similar to LERK-2 (59% amino acid identity), another ligand in this family that is a transmembrane protein [37, 47].
Cerretti et al. [46] used LERK-5 riboprobes to examine Northern blots of fetal and adult human tissues. A single mRNA transcript 5000 bp in length was observed in fetal tissues including heart, lung, kidney, and brain. Lung was the only adult tissue that had abundant levels of LERK-5 mRNA. As in humans, an mRNA transcript 5000 bp in length was present in 17.5-day bovine embryos, and preliminary evidence suggests that it was present primarily in adult bovine lung. Further studies are required to determine the significance of the marked increase in bLERK-5 mRNA content in bovine embryos between Days 15 and 17.5 of gestation.
Bovine IFN
Complementary DNA fragment 6 encoded bIFN, a protein secreted from the embryonic trophectoderm and responsible for maternal recognition of pregnancy in cattle [48, 49]. During luteolysis, prostaglandin F2
is released from the endometrium in a pulsatile manner causing the corpus luteum to regress [50]. Bovine conceptuses produce bIFN, which decreases prostaglandin F2 production by the endometrium and thus extends the lifespan of the corpus luteum, thereby stabilizing the uterine environment and allowing for continued growth of the embryo and its membranes [51]. Although maximal secretion of bIFN occurs between Days 16 and 19 of pregnancy [52], oIFN mRNA has been detected as early as Day 12 of gestation [53] and even earlier in bovine embryos produced in vitro [54].
Bovine IFN mRNA content increased 2.5-fold in 17.5-day relative to 15-day bovine embryos. It was surprising to find that bIFN mRNA was detected using six different primer combinations. Five oligo (dT) plus 2-bp primers were used to reverse transcribe and PCR amplify the cDNA products. A single decamer was used for PCR amplification that annealed specifically to the template 330 bp upstream from the 3' end of the bIFN cDNA [55]. After separating the reamplified products by gel electrophoresis on an agarose gel, a single band of 350 bp in length was present in all samples except cDNA fragment 10. Fragment 10 separated into two cDNA fragments approximately 600 bp and 350 bp in length. On the Southern blot, a single hybridization band was visible at 350 bp in all samples. A hybridization band was not present at 600 bp for cDNA fragment 10.
There are several plausible explanations for these results. Bovine IFN cDNA or the oligo (dT) plus 2-bp primer may have contaminated the PCR reaction components, thus explaining the presence of a hybridization band at 350 bp but not 600 bp for cDNA fragment 10. Also, there are several repeated adenine nucleotides dispersed throughout the 3' end of bIFN mRNA. Degenerate binding may have occurred between the oligo (dT) plus 2-bp primer and the mRNA transcript at the low annealing temperatures used in mRNA differential display analysis. This degeneracy likely would increase the possibility of detecting bIFN repeatedly using several primer combinations. One other possibility for bIFN is that different oligo (dT) plus 2-bp primers annealed to different members of this gene family [56].
IFN is the major translation product of mRNA isolated from Day 16 ovine [57, 58] and bovine conceptuses [59, 60]. The abundance of mRNA encoding bINF may increase the probability that degenerate hybridization will occur between the transcript and primer. Increasing the annealing temperature during PCR amplification would, in some cases, decrease recurrent isolation of the same cDNA fragment from multiple differential display analyses, but this must be balanced against failure to detect other differentially expressed genes.
Bovine PAF: A Novel Gene Product
Bovine PAF cDNA encoded a previously unidentified bovine gene with a mRNA transcript 1800 bp in length that increased 11-fold in 17.5-day relative to 15-day bovine embryos. The bPAF mRNA was present in the 25-day bovine cDNA library but declined to levels undetected by Northern analysis by 30 days of gestation, at least in extraembryonic membranes. There was a very small amount of tissue from embryonic disks in the 15.5- and 17.5-day embryos, but the fetus proper was discarded in the 30- to 36-day conceptuses. We therefore cannot rule out bPAF expression in the fetus itself. bPAF was not detected in bovine adult tissues by Northern blot analysis.
A single cDNA clone 1703 bp in length with a potential open reading frame of 381 bp encoding 126 amino acid residues was isolated from the cDNA library constructed from 25-day embryos. Neither the nucleotide sequence nor amino acid sequence demonstrated significant sequence homology to any previously reported transcripts. However, a similar DNA sequence with 83% nucleotide identity and 77% amino acid identity occurs in the human genome. This newly discovered gene merits further investigation because of its abundant expression during a narrow window of embryonic development and because of the intriguing properties of the protein specified by the open reading frame (Fig. 6). Bovine PAF could have an important developmental role, for example, as part of the cascade that leads to secretion of IFN.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: G.E. Seidel, Jr., ARBL, Colorado State University, Fort Collins, CO 80523. FAX: 970 491 3557; gseidel{at}colostate.edu
Received: 7 January 2003.
First decision: 29 January 2003.
Accepted: 8 May 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-amanitin sensitivity during bovine preattachment development. Mol Reprod Dev 2000 55:152-163[CrossRef][Medline]11 gene expressed in the bovine conceptus early in gestation. J Mol Endocrinol 1990 4:275-282 genes: relative transcript abundance during early placental development, promoter sequences of genes, and biological activity of protein products. Endocrinology 2001 142:2906-2915This article has been cited by other articles:
|
|
 |
|
  L. Blomberg, K. Hashizume, and C. Viebahn Blastocyst elongation, trophoblastic differentiation, and embryonic pattern formation Reproduction, February 1, 2008; 135(2): 181 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
X174/HaeIII
