| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Departments of Animal Sciences and
b Biochemistry, University of Missouri, Columbia, Missouri 65211
c Molecular Embryology Group, AgResearch-Ruakura, Hamilton, New Zealand
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) is secreted from trophectoderm of periimplantation ruminant conceptuses and is a critical component of pregnancy recognition. Multiple genes encode IFN-. The objectives of this study were to identify expressed forms of ovine IFN- and to compare their biological activities. Sequences analyzed after cloning 36 reverse transcription-polymerase chain reaction products of ovine conceptus RNA provided seven new cDNA that were similar in sequence to previously cloned forms (p3, p6, and p8 cDNA). Phylogenetic analysis of amino acid sequence for all new and previously reported forms showed that ovine IFN- forms can be divided into three main groups. Equivalent amounts of mRNA for p3, p6, and p8 forms were detected in conceptuses following RNase protection. Recombinant p3 and p8 protein had similar antiviral activity on ovine and bovine cells whereas p6 protein was less active. The p3 form was the most potent of the three in its ability to extend estrous cycle length in nonpregnant ewes. In summary, there appeared to be three main groups of ovine IFN-, each containing several variant forms. Antiviral activity was not particularly well correlated with ability to prevent luteolysis, suggesting that distinct intracellular mechanisms are used to exert the various actions of IFN-.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PGF2), thereby permitting continued CL function (see [1–4] for reviews). This antiluteolytic agent was initially named trophoblastin [5] or trophoblast protein-1 [6] but was later renamed interferon-tau (IFN-) when these genes were found to be structurally related to the type I IFN, which includes alpha, beta, and omega IFN [1]. The IFN- possess activities typical of all type I IFN; they protect cells against viral infection, regulate various aspects of immune cell function, and inhibit cell proliferation [7–12]. However, in contrast to other type I IFN, IFN- expression is not inducible by virus [13]. Rather, it is produced constitutively by the trophectoderm (tissue precursor of the placenta) during a limited period of pregnancy, e.g., Days 12–21 of pregnancy in sheep [14–16]. Genes encoding IFN- have been detected only in ruminants [17]. It is now surmised that IFN- evolved from an ancestral IFN-omega gene by a duplication event occurring approximately 36 million years ago, about the time of divergence of pecoran ruminants (sheep, cattle, antelope, deer, giraffe) from other artiodactyls (pigs, llamas, camels) [18, 19]. Simultaneously or soon thereafter, they acquired a trophectoderm-specific promoter and became a component of conceptus signaling [18, 19].
Strong evidence implicates IFN- as a major factor in maintenance of CL function during pregnancy in ruminants. Intrauterine administration of IFN- to nonpregnant sheep and cattle is sufficient to extend CL function for several days to a few weeks beyond its normal functional life span [6, 12, 20–27]. Such treatment does not extend CL function to that of a normal gestation length, however, indicating that IFN- serves as only the first in a series of signals required for pregnancy maintenance. The IFN- does appear to be more effective than IFN-alpha as an antiluteolytic agent [28, 29] and functions by preventing up-regulation of oxytocin receptors on endometrium during diestrus [30, 31]. This limits the ability of oxytocin to induce pulsatile release of PGF2, which is required for CL regression [20–22, 26, 32–34]. Failure of the endometrium to respond to the IFN- signal, and failure of the conceptus to produce sufficient quantities of IFN- at the required time, have been implicated as causes for early pregnancy failure in domestic ruminants [35].
Fifteen ovine (ov) IFN- gene sequences attained from sequencing either cDNA from libraries or genomic DNA have been entered into data banks [17, 36–40]. The IFN- genes, which are intronless, appear to differ both in their level of expression and in the biological activity of their protein products [12, 40, 41]. Previous work from this laboratory showed that the mRNA of two distinct ovIFN- forms, termed p3 and p6, were 10 times more abundant during early pregnancy than that of a third mRNA species, termed s4 [12]. In addition, p3 and p6 proteins extended CL life span more efficiently than s4 [12]. However, a full inventory of all expressed genes and the activities of their protein products have not been described, nor have the interrelationships of the different genes in the data banks been examined. Therefore, the work described here was aimed at distinguishing expressed genes from those that are likely pseudogenes and at assessing the biological activities of proteins from the expressed genes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents used included avian myeloblastosis virus (AMV) reverse transcriptase, deoxynucleotide triphosphates, RNA transcription kit, T4 DNA ligase, RNase-free DNase, restriction endonucleases, JM109 and BL21 Escherichia coli strains, pGEM-T and pGEM-T Easy vectors, and Wizard miniprep DNA purification system (Promega, Madison, WI); Taq polymerase, oligonucleotides, and Trizol reagent (Gibco/BRL, Gaithersburg, MD); Qiaex Agarose Gel Extraction kit (Qiagen, Santa Clarita, CA); isopropyl-ß-D-thiogalactoside (IPTG; Alexis, San Diego, CA); HybSpeed RNase protection assay kit (Ambion, Austin, TX); sequenase version 2.0 sequencing kit (U.S. Biochemical, Cleveland, OH); dispase (Boehringer-Mannheim, Indianapolis, IN); pGEX-2T vector and glutathione Sepharose (Pharmacia Biotech, Piscataway, NJ); Detoxi-Gel endotoxin removing gel (Pierce, Rockford, IL); Luria broth (Difco Laboratories, Detroit, MI); cell culture medium, medium antibiotics, and fetal bovine serum (University of Missouri Cell Culture Core Facility, Columbia, MO); mammalian cell lines (American Type Culture Collection [ATCC], Manassas, VA); [32P]CTP (600 Ci/mmol; DuPont/NEN, Wilmington, DE); [35S]dATP (> 1000 Ci/mmol; Amersham; now Amersham Pharmacia Biotech, Piscataway, NJ); XAR-5 and XRP film (Eastman Kodak, Rochester, NY); PGF2 (Lutalyse; Upjohn Co., Kalamazoo, MI); intravaginal progesterone release devices (CIDR; InterAg, Hamilton, New Zealand); Tygon microbore tubing (Fisher Scientific, Pittsburgh, PA); Coat-a-Count progesterone kit (Diagnostic Products, Los Angeles, CA). Pancreatin and all other reagents were purchased from Sigma Chemical Co. (St. Louis, MO).
Reverse Transcription and Polymerase Chain Reaction (RT-PCR)
Conceptuses were collected from crossbred ewes (Hampshire, Romanov, Rambouillet, and Columbia) at Days 12–13 (n = 10), Days 15–16 (n = 6), and Days 18–19 of pregnancy (n = 4) (Day 0 = day of breeding), and total cellular RNA was extracted by using the Trizol reagent. All RNA preparations were treated with RNase-free DNase for 30 min at 37°C and were then incubated at 75°C for 15 min to inactivate the enzyme. Each RNA preparation (2 µg) was reverse transcribed at 42°C for 60 min by using AMV reverse transcriptase, oligo(dT) primers, and 250 µM each of dATP, dCTP, dGTP, and dTTP. Thirty cycles of PCR (94°C for 1 min, 55°C for 1 min, 72°C for 1 min) were completed by using Gibco/BRL Taq polymerase and ovIFN--specific primers. The 5' primer (CATCTTCCCCATGGCCTTCG) corresponded to sequence surrounding the ovIFN- translation start site, and the 3' primer (GCATCTTAGTCGGCAAGAG) corresponded to the proximal 3' untranslated region of ovIFN-. Sequences corresponding to these primers are completely conserved among all known ovIFN- sequences. Non-reverse-transcribed total cellular RNA was included as a control for the presence of genomic DNA contaminants.
Complementary DNA Cloning and Sequencing
PCR products were agarose gel extracted (Qiaex gel extraction kit) and ligated into the pGEM-T vector by using T4 DNA ligase. Plasmids were used to transform JM109 E. coli, and cells were plated on Luria broth plates containing ampicillin (50 µg/ml), X-gal (40 µg/ml), and IPTG (1 mM). After blue/white colony selection, transformants were grown in Luria broth containing ampicillin (50 µg/ml) at 37°C overnight, and plasmid was isolated by using Wizard minipreps. Presence of inserts was verified by restriction digestion with SacI and ApaI. Thirty-six clones (twelve from each stage of development) were sequenced to completion by using the dideoxy method (Sequenase sequencing kit) with vector primers and two ovIFN--specific primers. Sequences were compared with those in the GenBank database, and novel sequences were sequenced again to verify their uniqueness.
RNase Protection Assays
Conceptuses derived either from Romanov x Romanov (n = 4) or Hampshire x Hampshire (n = 3) matings were recovered at Day 15 of pregnancy. The total cellular RNA was extracted from individual conceptuses with the Trizol reagent. Preparations were treated with RNase-free DNase for 30 min at 37°C.
OvIFN- templates corresponding to nucleotide positions +117 through +586 of the open reading frame for p3, p6, and p8 ovIFN- (GenBank accession nos. X56341, X56343, and X56345, respectively) were generated by using PCR. The 3' primer contained a T7 polymerase promoter sequence, and cRNA was synthesized from templates as described previously [12]. Specific activity for ovIFN- probes averaged 3.6 x 108 dpm/µg.
A bovine glyceraldehyde-3-phosphate dehydrogenase (boG3PDH; GenBank accession no. U85042) cRNA served as the internal RNA loading control. A partial cDNA (212 base pairs [bp]) was generated through RT-PCR of Day 150 bovine placental RNA and was cloned into the pGEM-T Easy vector. Exactness of sequence was verified by DNA sequencing. The plasmid was linearized with Nde1, and riboprobe was generated as described previously [12]. The bovine G3PDH cDNA differed from the ovine sequence by a single position and yielded a 197-base protected fragment following RNase protection with ovine RNA. Specific activity for the G3PDH probe was 1.0 x 108 DPM/µg.
RNase protection was completed as described previously [12] with 2 µg of total cellular RNA and 105 cpm of each riboprobe. Protected fragments were separated by electrophoresis in 5% (w:v) acrylamide gels containing 8 M urea. Dried gels were exposed to XAR-5 film for 3–18 h, and optical density units of bands were measured (GpTools, version 3.0; BioPhotonics, Ann Arbor, MI).
Collection and Culture of Primary Ovine Endometrial Epithelium
Reproductive tracts were retrieved from three ewes at Day 10 postestrus. The luminal epithelium was collected as modified from Bowen et al. [42]. In brief, 30 ml dispase (5 mg/ml) and pancreatin (0.125% [w:v]) in 0.01 M PBS (Ca/Mg free, pH 7.2) was injected into the uterine lumen, and tracts were incubated in a 37°C water bath for 60 min. The digestion solution was collected, and lumens were flushed with 60 ml 0.01 M PBS (Ca/Mg free, pH 7.2). Cells were centrifuged (1000 x g for 5 min) and washed by dispersion in Dulbecco's modified Eagle's medium (DMEM) containing 10% (v:v) fetal bovine serum, 100 IU/ml penicillin, 50 µg/ml streptomycin, and 2.5 µg/ml fungizone. Cells were passed through an 18-gauge needle to reduce tissue clump size and were incubated (37°C, 5.0% CO2 in air, 100% humidity) in DMEM containing supplements. Residual fibroblasts were removed by collecting the nonattached cells after 18 h in culture and re-plating. Cells were passaged with trypsin (0.05% [w:v] in Hanks' buffered saline solution) upon confluence (1–2 wk). Cells divided in culture and maintained a characteristic epithelial appearance of closely packed squamous cells when viewed by light microscopy for 8–11 passages. Visual morphology was the only criterion used for characterizing these cells.
Generation of Recombinant ovIFN-
Recombinant proteins for p3 (oatp1p3 in Fig. 2) and p6 (oatp1p6 in Fig. 2) were produced in E. coli as described previously [12]. The coding region of the first identified p8 cDNA (oatp1p8 in Fig. 2), which was derived from the same conceptus cDNA library as those used to generate p3 and p6 protein [38], was cloned into the pGEX-2T expression vector, and protein was generated [12]. All preparations were passed through a Detoxi-Gel column to remove endotoxic contaminants; they were then sterile filtered and stored at 4°C. Each recombinant protein preparation was similar in purity when assessed visually after polyacrylamide gel electrophoresis and Coomassie staining (data not shown).
|
Antiviral Assays
Assays were completed as described previously [7] on Madin-Darby bovine kidney cells (MDBK; ATCC #CCL-22), Big-Horn fetal tongue epithelium cells (BHFTE; ATCC #CRL-1941), and primary cultures of ovine endometrial epithelium. Cells were exposed to 3-fold serial dilutions of each protein. After 24 h, cells were challenged with vesicular stomatitis virus (VSV), and cells were stained with gentian violet 24 h later. A laboratory standard (boIFN-1; 5.4 x 107 IU/mg) was used to determine specific activity of proteins on MDBK cells. No standards were available for the ovine cells. Therefore, the protein concentration that provided a 50% inhibition in cytopathic protection assays (with VSV as the challenge virus) was used to define antiviral activity on these cells.
Intrauterine Injection of ovIFN-
Two separate intrauterine injection studies were completed as described previously [6, 12]. The first study was done at the Ruakura Experimental Station of AgResearch in Hamilton, New Zealand. Estrus was synchronized in Romney and Coopworth ewes (n = 41) by intravaginal insertion of a CIDR-G device. The CIDR devices were removed 14 days later, and ewes were detected in standing estrus by using vasectomized rams. On Days 5–7 postestrus, ewes were fitted with indwelling uterine catheters. Ewes within each breed were assigned at random to receive either 10 or 100 µg/day of p3, p6, or p8 ovIFN- or a control treatment (PBS containing 0.1% [w:v] ovine albumin) from Days 11 to 15 postestrus. From 20 to 200 µg/day of ovIFN- is secreted daily from individual ovine conceptuses at Days 14–16 of pregnancy [43]; therefore the amounts introduced into ewes in these experiments were within the physiological range. Sterile protein solutions (2 ml; 0.01 M PBS, pH 7.2 containing 0.1% [w:v] ovine albumin) were introduced into the uterus twice daily, with half the daily dose given at 0700 and 1900 h. Catheters were flushed after each treatment with 2 ml PBS containing 0.1% (w:v) ovine albumin, 50 IU/ml penicillin G, and 25 µg/ml streptomycin. To evaluate serum progesterone profiles, blood samples were collected on the day of estrus, at Day 7 postestrus, and every other day from Day 10 to 20 postestrus. Catheters were removed surgically on Days 19–20 postestrus, and ovaries were assessed for recent ovulation. Blood sampling continued for ewes that had not undergone a recent ovulation until they returned to estrus or until Day 30 postestrus.
The second study was completed at the University of Missouri. Procedures used were identical to those for the first study with the following exceptions. Estrus of Ramboillet/Columbia crossbred ewes (n = 32) was induced by two i.m. injections of PGF2 (Lutalyse; 15 mg) given 11 days apart. Surgical procedures were modified so catheters could be maintained longer. The ovary contralateral to the horn containing the catheter was removed to ensure that ovulation would occur ipsilateral to the catheter-bearing horn during subsequent estrous cycles. The external portion of the catheter was run subcutaneously from the flank to the back, where it was exteriorized, sutured in place, and protected with a water-resistant patch. Ewes were assigned randomly to receive either 50 or 500 µg/day of p3, p6, or p8 ovIFN- from Days 11 to 15 postestrus. A control treatment was run simultaneously. Blood samples were collected as described previously either until ewes were observed in standing estrus or until Day 30 postestrus.
Progesterone Assays
Concentration of progesterone in serum was determined by using the Coat-A-Count RIA kit. Assay sensitivity was 0.02 ng/ml, and intraassay coefficients of variation were 15.6% and 9.5% for the first and second intrauterine injection study, respectively. Life span of the CL was defined as the number of days from the synchronized estrus to the subsequent fall in serum progesterone (< 0.5 ng/ml).
Statistical Analysis
Amounts of ovIFN- mRNA detected by using RNase protection were determined by normalizing optical density units relative to amount of G3PDH mRNA. Optical density units were also corrected for fragment size. Differences between ovIFN- form and breed were contrasted by using least-squares analysis of variance (LS-ANOVA) [44], and differences between individual means were partitioned further by using the Duncan test [44].
Differences in antiviral activity between ovIFN- proteins were determined within each cell type by using LS-ANOVA [44] after data were log transformed. Differences between individual means were evaluated further by the Duncan test [44].
To evaluate differences in CL life span with different dosages of ovIFN-, values were transformed by nonparametric ranking, and each study was analyzed by LS-ANOVA [44]. Differences between means were partitioned further by using the Duncan test [44].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Products of the expected size (615 bp) were generated by RT-PCR at each stage of conceptus development examined (Fig. 1). These amplified products were shown to be derived from reverse-transcribed DNA and not from genomic DNA by completing parallel reactions with total cellular RNA that had not been reverse transcribed. In no cases were bands observed in these controls.
|
Of the 36 clones sequenced (12 from each stage of development), 31 were either identical to or differed in only a few bases from p8 ovIFN-. The four new sequences are depicted in Figure 2 and are named otaup8v1, otaup8v2, otaup8v3, and otaup8v4 (GenBank accession nos. AF158817 to AF158820). One of the remaining five clones was identical to p6 ovIFN-. The other four differed by only a few bases from p6 and are named otaup6v1, otaup6v2, and otaup6v3 (GenBank accession nos. AF158821 to AF158823) (Fig. 2). No other forms were identified by this method, not even the p3 form, which was shown to be expressed at levels comparable to those for p6 in a previous study [12]. Such bias toward cloning p8 forms was surprising, since the ovIFN- transcripts being amplified were of identical length and similar sequence.
Alignment of ovIFN- Sequences
Both sequence alignments (Fig. 2) and an evolutionary phylogram (Fig. 3) for the new and previously identified ovIFN- were created. The newly identified sequences are depicted in bold type on Figure 2. There were two instances in which multiple cDNA encoded the same protein. A series of four cDNA (oatp1p7, oatp1p8, shptp, and osotp1; GenBank locus names) [37–39] encoded the same protein, which is represented by the oatp1p8 sequence in Figures 2 and 3. Two other cDNA (oatp1p6 and oatp1p12) [37] encoded the same protein, which is represented by the oatp1p6 sequence (Figs. 2 and 3). Three classes of ovIFN- could be distinguished based on the phylogram (Fig. 3). Forms within each class differed by no more than 7.0% from each other in amino acid sequence. Forms in different classes differed from 5.2% to 13.4% in amino acid sequence. Both p3 and p8 are categorized as class A forms, whereas p6 represents a class B form.
|
More forms of ovIFN- could be detected in class A than in either class B or C. Class B and C sequences contained a site for glycosylation at Asn 78 (Fig. 2). However, glycosylated forms of ovIFN- have not been reported [45, 46].
Messenger RNA Abundance of ovIFN- Forms
RNase protection assays were completed to compare transcript abundance between the main forms of ovIFN- identified in this and previous studies (p3, p6, and p8) [12]. Transcripts for all three forms were detected from conceptus RNA derived from both Romanov and Hampshire intrabreed matings, as reflected by the presence of full-length protected fragments following RNase digestion (Fig. 4). There was no breed effect on quantity of ovIFN- mRNA. No differences in mRNA abundance were detected between p3, p6, and p8 (Fig. 5A).
|
|
Protected fragments of smaller size were also detected with each ovIFN- probe (Fig. 4). These fragments did not result from the protection of partially transcribed probes, since no such fragments were detected in undigested controls (Fig. 4). Rather these fragments likely represent mRNA populations that differed from p3, p6, or p8 by only a few bases. For example, two partially protected p6 fragments (approximately 260 and 200 bases in size) probably reflect the presence of p12 mRNA, which differs from p6 at a single nucleotide position [38].
A partially protected fragment of ~350 bases was seen with use of the p8 probe and to a lesser extent with use of the p3 probe in three of the four Romanov conceptuses (Fig. 4). This fragment was absent in all of the Hampshire conceptuses (Fig. 4). The identity of this mRNA species is unknown, but the species is presumably a class A form. Some of the other partially protected fragments varied in their intensity between individual conceptuses, but this variation was not associated with breed.
To assess differences in the relative abundance of mRNA between the two classes of expressed ovIFN- (A and B), the total quantities of both full-length and partially protected fragments above 100 bases in length were summed together. Total mRNA abundance did not differ between class A (p3 and p8) and B (p6) mRNA (Fig. 5B).
Antiviral Activities of ovIFN- Proteins
All recombinant proteins had antiviral activity (Table 1), and their activities on MDBK cells were within the range of activities reported previously [9, 11, 12, 27]. Both p3 and p8 possessed greater (P < 0.05) antiviral potencies than p6 on all cells tested. The difference in potency between p6 and other proteins was more pronounced on ovine cells than on the bovine cell line.
|
Effect of Intrauterine Injection of ovIFN- Proteins on Estrous Cycle Length in Nonpregnant Ewes
In the first study (Fig. 6A), control ewes had CL life spans of normal duration. None of the proteins significantly extended CL life span at a dosage of 10 µg/day. The p3 protein increased (P < 0.05) life span of the CL at a dosage of 100 µg/day, whereas no CL extension was detected with the same dosage of p6 and p8.
|
In the second intrauterine injection study (Fig. 6B), seven of the ewes, including three controls, had short cycles (8–12 days), did not display behavioral estrus, and showed no subsequent rise in serum progesterone. Because this study was completed late in the breeding season (February/March), these ewes were considered to have undergone seasonal anestrus and were excluded from analysis. The remaining control ewes had CL life spans of normal duration. Life span of the CL was not significantly extended when proteins were given at a dosage of 50 µg/day, although some ewes exhibited an estrous cycle extension. Both p3 and p8 increased (P < 0.05) CL life span when given at a dosage of 250 µg/day. By contrast, CL life span of ewes given 250 µg/day of p6 did not differ from the control value. However, a previous study had shown that a similar dosage of p6 (300 µg/day) was effective at extending CL life span in ewes [12]. The inability to detect a significant CL extension with 250 µg/day of p6 in the present study was likely the result of an insufficient number of ewes in this treatment group, since two of the four ewes initially assigned to this treatment had short estrous cycles and became anestrous. Antiviral activities of all protein preparations did not change during either intrauterine infusion study, indicating that effects observed were not caused by variation in the stability of proteins.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
transcripts and genes. Most notably, Klemann et al. [38] identified six distinct cDNA from a Day 15 to 16 ovine conceptus cDNA library, and Nephew et al. [40] identified five forms by genomic cloning. The RT-PCR approach used in this study provided seven additional expressed forms of ovIFN-, which now increases the total number of different ovIFN- sequences to 22. However, all sequences acquired in this study differed at only a few base positions from the ones already described. It has not yet been established whether mRNA species that represent these new sequences are expressed or whether these novel sequences resulted from mutations introduced during RT-PCR. It remains unclear how many of the sequences in Figure 2 represent distinct genes and how many represent allelic polymorphisms.
A majority of sequences identified from the RT-PCR-based cloning strategy encoded p8 or p8-like forms, but it was clear from RNase protection that p8 mRNA was no more abundant than either p3 or p6 mRNA. Such disparity was probably caused by subtle differences in p8 structure that allowed it to be more readily reverse transcribed or PCR amplified than other forms. This bias in RT-PCR amplification of mRNA has implications that go beyond this work because it raises the possibility that this event may be a common phenomenon that could undermine both quantitative and semiquantitative analyses of mRNA in general. Because this bias in RT-PCR-based amplification existed, it remains unclear whether all of the expressed forms of ovIFN- have been identified.
All the known expressed forms of ovIFN- reside within class A and B, whereas class C forms likely represent pseudogenes or poorly expressed forms. All class C sequences were obtained through genomic cloning [17, 40], and none have been detected as cDNA. One of these, shpotp1c (also termed s4), appeared to be expressed at low levels by the ovine conceptus [12, 41]. Another class C form, shp02tp, was reportedly expressed poorly relative to other ovIFN- transcripts, but the RT-PCR procedure used to make this assessment was not quantitative [40]. The remaining form, shpotp1b, encodes a 161 amino acid protein [17]. No such truncated ovIFN- proteins have ever been detected from ovine conceptus secretions [6, 36, 37, 45, 46]. Interestingly, gene promoters of class C variants can be distinguished from those more abundantly expressed ovIFN- forms by a mutation in a core sequence that is necessary for IFN- gene transactivation by Ets-2 [41].
All but three class A and B sequences were derived from cDNA [36–39]. Nephew et al. [40] concluded that one class A form cloned from genomic DNA (shp010tp) was the predominant form of ovIFN- mRNA expressed by ovine conceptuses. However, the shp010tp-specific primer used was clearly not specific and would have probably amplified all class A mRNA.
Based on RNase protection, class A and B transcripts seem to contribute equally to the total ovIFN- mRNA in the ovine conceptus. However, mRNA populations measured were limited to those three being studied (p3, p6, and p8 variants). Thus, the total mRNA concentrations were probably underestimated, particularly within class A, where some mRNA species possess as many as 10 nucleotide differences from p3 and p8 and would have been degraded more or less completely by the RNase.
It was previously determined that p3 and p6 mRNA abundance was similar throughout periimplantation conceptus development [12]. However, it remains unknown whether other mRNA populations, such as p8, exhibit similar temporal expression patterns or whether some ovIFN- forms are preferentially expressed earlier or later during conceptus development.
As anticipated [47–49], the RNase protection assay seemed capable of detecting single base mismatches. With the p6 riboprobe, for example, partially protected fragments of 260 and 200 bases in length were noted. These fragments matched the lengths that were expected if another class B form, p12, that differs in a single base position from p6 had been hybridized to the probe [38]. One partially protected fragment was detected only in Romanov conceptus RNA with use of the p8 probe, and to a lesser extent with the p3 probe. Such breed-associated polymorphisms were not unexpected and could represent either an allelic or a distinct gene variant form of p8.
It was of interest to establish whether distinct class A forms differed in biological activity, despite their close structural similarities, and also to determine whether their bioactivities differed from those of class B forms. The p3 protein, a class A form, had previously been shown to be a potent antiluteolytic agent [12, 27]. The present work indicated that it was more potent than the other ovIFN- tested, even in comparison with the p8 protein, which differed by only three amino acids (G101, E107, Y128). Perhaps the presence of a tyrosine at position 128 of p8 in place of the histidine residue found in p3 is responsible for its slight reduction in antiluteolytic activity, since this residue is in close proximity to a putative receptor binding region [18, 50]. In any case, p3 must contain some structural feature that is preferred for activation of the intracellular signaling systems that modulate antiluteolytic activity within the uterus. The reduced antiluteolytic potency of p6 may have been expected since its antiviral activity was considerably less than that of the others; however, a direct relationship between these two activities remains to be addressed critically.
Some ewes did not exhibit extensions in interestrous intervals when given a dosage of ovIFN- sufficient for a response in their flock mates. It is not uncommon to observe ewes that are unresponsive to ovIFN- treatment [6, 12, 20–27], and such insensitivity to ovIFN- could be a cause of pregnancy failure in sheep and cattle [35]. Other ewes did respond to low dosages of ovIFN- (10–50 µg/day), and it is tempting to speculate that these ewes would have been better able to maintain a pregnancy than their flock mates had they been inseminated.
The antiviral potencies of the ovIFN- on ovine cells did not correlate well with their ability to extend CL life span. The p3 and p8 proteins had equivalent antiviral activities but differed in their antiluteolytic activities. In addition, in a previous study [12] the p6 protein was able to extend CL life span at a dosage (300 µg/day) that was similar to the dosage required for p8, but p6 was a less potent antiviral agent, particularly on ovine cells. Therefore, this suggests that the signaling systems that lead to antiluteolytic activity and antiviral activity may well be distinct.
It remains unclear why numerous ovIFN- forms are expressed. Perhaps expression of multiple genes ensures that sufficient quantities of protein will be present when needed. Even though large quantities of ovIFN- (20–200 µg/day) are secreted at Days 14–16 of pregnancy, very little, by comparison, is secreted at Day 12 of pregnancy (approximately 1–2 µg/day) [43], when the pregnancy recognition signal must first be realized [5, 51]. It is also possible that multiple ovIFN- proteins are required because they differ in their ability to influence other events associated with pregnancy, such as regulating the local immune system and secretion of pregnancy-specific uterine proteins. It may also be that ovIFN- proteins may act synergistically in some manner to achieve their required effects on the uterus.
|
ACKNOWLEDGMENTS |
|---|
protein alignment and growth tree; Mr. John Bader and Mr. John Lange for assistance in sheep studies; and Dr. Jonathan Green, personnel in Dr. Michael Roberts' laboratory, and personnel at AgResearch for assistance in surgical procedures. G.L.W. received salary support from the Howard Hughes Undergraduate Research Internship Program. |
FOOTNOTES |
|---|
2 Correspondence: Alan D. Ealy, Department of Animal Sciences, 158 Animal Science Research Center, University of Missouri, Columbia, MO 65211–5300. FAX: 573 882 6827; ealya{at}missouri.edu
Accepted: August 10, 1999.
Received: May 17, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(IFN-): changes in receptor binding and in antiviral and antiproliferative activities resulting from site-directed mutagenesis performed near the carboxyl terminus. J Biol Chem 1994; 269:24826–24833. and E and of 13,14-dihydro-15-keto prostaglandin F-2 in cyclic ewes. J Reprod Fertil 1988; 84:493–504.-interferon (trophoblastin) in sheep. J Endocrinol 1990; 127:R5-R8. extends corpus luteum lifespan in cyclic ewes. Theriogenology 1995; 43:1031–1045. of cows in response to recombinant interferon-. J Dairy Sci 1995; 78:1921–1931.[Abstract] or TP-1 or trophoblastins) in the maternal recognition of pregnancy. In: Glasser SR (ed.), Endocrinology of Embryo-Endometrium Interactions. New York: Plenum Press; 1994: 195–216. regulates expression of endometrial receptors for estrogen and oxytocin but not progesterone. Biol Reprod 1995; 53:732–745.[Abstract] by intrauterine and intramuscular administration of recombinant bovine interferon- to cows. J Reprod Fertil 1991; 93:375–384.1-1 decrease endometrial oxytocin receptor concentrations in cyclic and progesterone-treated ovariectomized ewes. J Endocrinol 1991; 131:475–482. genes encoding functional class II and trophoblast (oTP-1) -interferons. Mol Cell Endocrinol 1991; 76:161–171.[CrossRef][Medline] gene expression by Ets-2. Proc Natl Acad Sci USA 1998; 95:7882–7887. at 2.1 angstrom resolution. J Mol Biol 1999; 286:151–162.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  Y. Chen, J. A. Green, E. Antoniou, A. D. Ealy, N. Mathialagan, A. M. Walker, M. P. Avalle, C. S. Rosenfeld, L. B. Hearne, and R. M. Roberts Effect of Interferon-{tau} Administration on Endometrium of Nonpregnant Ewes: A Comparison with Pregnant Ewes Endocrinology, May 1, 2006; 147(5): 2127 - 2137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.P. Green, L.D. Spate, J.A. Bixby, A.D. Ealy, and R.M. Roberts A Comparison of the Anti-Luteolytic Activities of Recombinant Ovine Interferon-Alpha and -Tau in Sheep Biol Reprod, December 1, 2005; 73(6): 1087 - 1093. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-I. Kawamoto, K. Oritani, H. Asada, I. Takahashi, J. Ishikawa, H. Yoshida, M. Yamada, N. Ishida, H. Ujiie, H. Masaie, et al. Antiviral Activity of Limitin against Encephalomyocarditis Virus, Herpes Simplex Virus, and Mouse Hepatitis Virus: Diverse Requirements by Limitin and Alpha Interferon for Interferon Regulatory Factor 1 J. Virol., September 1, 2003; 77(17): 9622 - 9631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Parent, C. Villeneuve, A. P. Alexenko, A. D. Ealy, and M. A. Fortier Influence of Different Isoforms of Recombinant Trophoblastic Interferons on Prostaglandin Production in Cultured Bovine Endometrial Cells Biol Reprod, March 1, 2003; 68(3): 1035 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Ealy, S. F. Larson, L. Liu, A. P. Alexenko, G. L. Winkelman, H. M. Kubisch, J. A. Bixby, and R. M. Roberts Polymorphic Forms of Expressed Bovine Interferon-{{tau}} Genes: Relative Transcript Abundance during Early Placental Development, Promoter Sequences of Genes and Biological Activity of Protein Products Endocrinology, July 1, 2001; 142(7): 2906 - 2915. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Swasdipan, H. Bielefeldt-Ohmann, N. Phillips, P. D. Kirkland, and M. R. McGowan Rapid Transplacental Infection with Bovine Pestivirus Following Intranasal Inoculation of Ewes in Early Pregnancy Vet. Pathol., May 1, 2001; 38(3): 275 - 280. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
