| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Genes: Relative Transcript Abundance during Early Placental Development, Promoter Sequences of Genes and Biological Activity of Protein Products1
Department of Animal Sciences, University of Missouri, Columbia, Missouri 65211
Address all correspondence and requests for reprints to: Dr. R. Michael Roberts, Department of Animal Sciences, University of Missouri, 158 Animal Science Research Center, Columbia, Missouri 65211-5300. E-mail: robertsrm{at}missouri.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
genes exist in cattle, but it has
remained unclear how many are expressed, the extent of their variation,
and whether different genes exhibit similar patterns of expression and
code for proteins with similar biological activities. A total of 118
complementary DNA (cDNA) were bi-directionally sequenced from
reverse-transcribed bovine (bo) conceptus RNA over the period from
blastocyst formation until day 25 of pregnancy. Fourteen
different cDNAs, encoding eight different IFN-, were confirmed
unique. All showed high sequence conservation (>98% nucleotide
identity; >96% amino acid identity). The cDNA fell into three,
recently evolved, phylogenetic groups (1, 2, and 3). Mean
concentrations of IFN- messenger RNA were greater at day 17 and day
19 than at day 14 and day 25, with different genes showing
comparable expression patterns, although there appeared to be a major
bias in expression of two genes (for boIFN-1c and 3a) in
blastocysts. Genes representing members of the three boIFN- groups
were cloned. Their promoter regions were conserved over regions
considered important for transcriptional activation. Recombinant
protein generated in Escherichia coli from
representative genes in the three groups had similar but not identical
antiviral activities. In summary, many IFN- genes, which are
probably under similar transcriptional control, are expressed in bovine
trophoblast during the peri-implantation period of development. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
are members of the
Type I IFN, an extensive grouping that includes the IFN-
, -ß, and
-
(1). Although the IFN- possess many of the
functional features of other Type I IFN, such as the ability to prevent
viral infection and to limit cell proliferation, they are unique in
that they are not induced by viral infection and are not widely
expressed (2). The production of IFN- is restricted to
the preimplantation period of conceptus development in ruminant
ungulates, when they are secreted between the blastocyst stage and the
time when the elongated trophoblast makes definitive attachment to the
uterine wall (2, 3, 4). The function of the IFN- is to
prevent the destruction of the corpus luteum, a process that would
normally occur at the end of a nonfertile estrous cycle from the
pulsatile release of prostaglandin F2
(PGF2) from the uterine endometrium (2, 5, 6). The current view is that IFN- acts on the endometrium
to abrogate production of luteolytic pulses of
PGF2 by reducing expression of estrogen and
oxytocin receptors in uterine epithelial cells (7, 8).
Others have suggested that exposure of uterine epithelial cells to
IFN- leads to transcriptional inactivation of the Cox-2
gene (9).
Based on base substitution rates, it has been calculated that the first
IFN- gene (IFNT) originated by a single duplication event
from an IFN- gene (IFNW) about 36 million years ago in
the mammalian lineage, leading to the present day pecoran ruminants, a
suborder comprised of cattle, deer, giraffes and their relatives
(1, 10). It has been speculated that the initial
duplication event that provided the primordial IFNT also
disrupted the promoter element, which ultimately led to restricted
trophoblast expression and to this novel role in maternal recognition
of pregnancy (1). The genes have continued to duplicate
since then, and it has been estimated that there may be as many as ten
IFNT in cattle, with all of them clustered within or in
close proximity to the genetic locus that contains the other Type I IFN
genes (11). Exactly how many IFNT genes are
expressed is unknown. Until this report, the sequences of only four
closely similar bovine (bo) IFN- cDNA sequences had been deposited
in GenBank (12, 13, 14). By contrast, many distinct
transcripts and genes have been reported for sheep IFN-
(15, 16, 17, 18, 19, 20, 21). Two-dimensional PAGE has indicated that at least
four isoelectric variants and several size variants of boIFN- are
secreted by cultured conceptuses (22, 23, 24). Some of this
heterogeneity is due to the addition of carbohydrate (23, 24), but it also seems likely that more than a single gene is
transcribed.
The main objectives of this work were to identify the predominant
transcripts encoding IFN- in bovine conceptuses from a wide range of
breeds and to determine whether there are changes in the relative
messenger RNA (mRNA) abundance of the identified boIFN- forms as
conceptus development proceeds. Effort was also directed toward
determining the relative biological activities of protein products of
these genes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RNA was isolated from bovine conceptuses either by using the guanidium thiocyanate/cesium chloride method (26) or with the RNA Stat-60 reagent (Tel-Test, Inc., Friendswood, TX). All RNA preparations were treated with RNase-free DNase (Promega Corp., Madison, WI) for 15 min at 37 C, and then were heated to 75 C for 15 min to denature the DNase.
RT-PCR and cloning cDNA
Samples of total cellular (tc)RNA (1 µg) isolated from the
blastocyst-stage pool (n 
450), a day 17 conceptus pool
(n = 6), and a single d 19 conceptus were reverse transcribed for
1 h at 42 C by using AMV RT (Promega Corp., Madison,
WI), oligo dT primer, and 250 µM each of dATP, dCTP,
dGTP, and dTTP (Promega Corp., Madison, WI). Thirty cycles
of PCR were completed (95 C 15 sec, 56 C 15 sec, 72 C 40 sec) by using
KlenTaq polymerase (AB Peptides, St. Louis, MO) in combination with
proofreading polymerase (P.F.U.; Stratagene, La Jolla,
CA). The 5' primer (CATCTTCCCCATGGCCTTC) overlapped the 5' untranslated
region and transcriptional start site, whereas the 3' primer
(CATCTTAGTCAGCGAGAGTC) corresponded to the proximal 3' untranslated
region. These sequences are conserved among all known boIFN-
cDNA.
A single dATP overhang was incorporated into the PCR products by
incubation with 1 mM dATP and Taq polymerase
(Life Technologies, Inc., Gaithersburg, MD) at 70 C for 30
min. PCR products were then ligated into the pGEM-T Easy vector
(Promega Corp., Madison, WI) by using T4 DNA ligase
(Promega Corp.) and were used to transform JM109
Escherichia coli (Promega Corp.). Selected
clones that contained inserts of the proper size were sequenced
completely in both directions by using the dideoxy method with vector
primers and boIFN--specific internal primers.
A second round of RT-PCR amplification was carried out identically to the one described above on RNA from one batch of 100 day 7 blastocysts, two batches of (n = 79 and n = 46, respectively) day 9 blastocysts, and from two groups blastocyst outgrowths (n = 45 and n = 43, respectively) collected 14 days after blastocyst formation. As in the first experiment, the oocytes were predominantly from beef cows, fertilized with mixed semen from six Holstein bulls (25).
The fidelity of PCR was determined to assess whether the appearance of
distinct cDNAs was a true reflection of the existence of different mRNA
molecules or resulted from mutations incorporated into templates during
PCR. Four independent PCRs were completed with a single template
(1c; clone d25–4; see Table 1
) and
the same primers, enzymes, and conditions discussed previously.
Products were cloned into the pGEM-T easy vector as described and a
total of ten cDNAs (from one to four cDNAs from each individual PCR)
were sequenced.
|
32P]-dATP-labeled (3000 Ci/mmol, NEN Life Science Products, Wilmington, DE) partial cDNA fragment of
boIFN-1a (14). Selected positive plaques were
identified, and the inserts amplified in plasmid. The cDNA were
sequenced by using vector primers and internal primers that were
specific for boIFN-.
RNase protection assays
To generate DNA templates for cRNA production, DNA inserts of
boIFN-1a, 2b and 3d (GenBank Accession Nos. M31557, AF196322
and AF196327, respectively) were generated by using PCR and subcloned
into the pGEM-T Easy vector. The sequences of the subcloned products
were confirmed. The DNA inserts were 325 bp in length and corresponded
to nucleotide positions +167 to +491 of the open reading frame. These
templates differed in seven to ten base positions from each other and
by at least one base position from all other forms. A partial cDNA (212
bp) for bovine glyceraldehyde-3-phosphate dehydrogenase (boG3PDH) was
used to control for RNA loading (21).
Riboprobe synthesis and RNase protection was completed as described
earlier (28). In brief, all complementary RNA were
transcribed with an in vitro transcription kit
(Promega Corp., Madison, WI) in presence of 100 µCi
[32P]-ribose CTP (600 Ci/mmol; NEN Life Science Products, Wilmington, DE) and 12
µM ribose CTP and 500
µM ribose ATP, GTP and UTP. Specific activity
averaged 1.9 x 108 DPM/µg for boIFN-
probes and 1.0 x 108 DPM/µg for
boG3PDH.
The RNase protection assays were completed according to manufacturer’s
instructions (Hybspeed RPA; Ambion, Inc., Austin, TX) with
5 µg of total cellular RNA from each conceptus preparation and
105 cpm of each riboprobe. Protected fragments
were separated by electrophoresis in 5% (wt/vol) acrylamide gels
containing 8 M urea. Dried gels were exposed to XAR-5 film
(Eastman Kodak Co., Rochester, NY) for 6 to 24 h, and
optical density units of bands measured (GpTools, version 3.0;
BioPhotonics Corp., Ann Arbor, MI). The amount of fully protected
boIFN- mRNA was normalized relative to boG3PDH mRNA. Assays were
replicated three times. Differences in mRNA abundance between variants,
day of conceptus development, and their interaction were determined by
using least squares ANOVA (LS-ANOVA) (29). Differences
between individual means were partitioned further by pair-wise
comparisons for probability of individual differences
(29).
Genomic boIFNT cloning
To examine the promoter regions of boIFNT, calf liver
DNA (Roche Molecular Biochemicals, Indianapolis, IN) was
used as template for thirty cycles of PCR (95 C 15 sec, 56 C 15 sec, 72
C 90 sec) with KlenTaq and P.F.U. polymerases. The primers used
corresponded to conserved sequences in IFNT that were
approximately 400 bases upstream of the transcription start site (5'
primer: AATACAAACATCAATATGGCC) (19) and within the
proximal 3' untranslated region of boIFNT, approximately 690
bases from the transcription start site (3' primer:
CATCTTAGTCAGCGAGAGTC). The amplified products were cloned into the
pGEM-T Easy vector. Twelve clones that contained inserts of the correct
size were sequenced. Promoter regions were sequenced in both directions
with vector and internal primers, and coding regions were sequenced
once to determine which class (T1, T2 or
T3) they represented.
Recombinant protein production
Recombinant boIFN-1a (also termed bTP509) was produced in
Escherichia coli as described previously (30).
Three additional recombinant proteins were generated by cloning the
coding regions of 1c, 2b, and 3b into the pET11a bacterial
expression vector (Novagen Inc., Madison, WI). The inserted sequences
were generated by using PCR with KlenTaq and P.F.U. polymerases. The
resulting DNA fragments contained an NdeI restriction site
at the 5' end, an N-terminal methionine codon, the boIFN- coding
region sequences minus their signal peptide sequences, and a
BamHI restriction site at the 3 ' end. PCR products were
first cloned into the pGEM-T easy vector, as described earlier, and
then were digested with NdeI and BamHI to release
the insert. The digested inserts were ligated into NdeI and
BamHI-digested pET11a vector with T4 DNA ligase and were
amplified in JM109 Escherichia coli. Proper insertion of the
cDNAs and integrity of their sequences were determined by DNA
sequencing. The clones were then used to transform BL21 pLysS DE3 Codon
Plus Escherichia coli (Promega Corp., Madison,
WI). Selected colonies were grown at 37 C in Luria Broth containing 50
µg/ml Ampicillin and 17 µg/ml chloramphenicol. At the appropriate
cell density (OD600 = 1), protein expression was
induced by addition of 0.5 mM isopropyl-ß
-day -thiogalactoside (IPTG; Alexis Corp., San
Diego, CA). Six hours later, cells were harvested and incubated in
bacterial protein extraction reagent (B-PER; Pierce Chemical Co., Rockford, IL). The insoluble protein fractions (inclusion
bodies) were collected by centrifugation (10,000 x g
for 15 min.) and dissolved in 6 M guanidinium-HCl
in 20 mM Tris (pH 8.0) containing 0.1%
[vol/vol] 2-mercaptoethanol. Protein refolding was completed as
described previously (30). In brief, 5 volumes of 20
mM Tris-HCl buffer (pH 8.0) containing 0.1%
[vol/vol] 2-mercaptoethanol was added drop-wise over 4 h. The
solutions were then dialyzed against 10,000 volumes of 20
mM Tris (pH 8.0), sterile-filtered and stored at
4 C.
Electrophoresis and western blotting
Proteins were analyzed by electrophoresis in 12.5% [wt/vol]
polyacrylamide gels containing 0.1% [wt/vol] SDS. Gels were either
stained with Coomassie dye (GelCode Blue Reagent; Pierce Chemical Co., Rockford, IL) or were subjected to western blotting. The
antiserum used was generated by immunizing rabbits with recombinant
boIFN-1a (bTP509) (30) and was diluted before use
(1:5000 [vol/vol]) in 10 mM Tris-HCl buffer (pH 8.0), 150
mM NaCl, 0.05% [wt/vol] Tween-20, and Escherichia
coli protein extract (0.5 mg/ml). Goat antirabbit IgG alkaline
phosphatase conjugate was used to detect bound immunoglobulins.
Antiviral assays
Antiviral assays were completed as described previously
(31) on Madin-Darby bovine kidney cells (MDBK;
ATCC#CCL22) (n = 7 assays) and bovine endometrial
cells (BEND; ATCC#CRL-2398; generously provided by T.
R. Hansen, University of Wyoming, Laramie, WY) (n = 6
assays). Cells were exposed to 3-fold serial dilutions of each
boIFN- in culture medium that contained 10% [vol/vol] FBS. After
24 h, cells were challenged with vesicular stomatitis virus and
were stained with gentian violet 19 h later. The antiviral
activities on MDBK were calculated by using a laboratory standard
(boIFN-1a; 5.4 x 107 IU/mg) that had
been standardized against a human IFN- reference reagent provided by
the Antiviral Substance Program of NIAID, NIH (Bethesda, MD). The
concentration of boIFN- that provided 50% inhibition in
virus-induced cell death was determined for both MDBK and BEND cells.
Differences in antiviral activities and potencies between boIFN-
forms were determined by using LS-ANOVA (29) after data
had been log-transformed.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
forms cDNA, i.e. an
error rate of 0.092%. Errors occur sufficiently frequently that single
base substitutions that are not observed in more than one PCR must be
viewed with caution. Therefore, we have chosen to report in this paper
only those nucleotide substitutions that have been observed in at least
two different amplification reactions.
The first experiments reported here identified 24 different cDNA from
48 clones from a variety of Bos taurus breeds. Each cDNA was
fully sequenced in both directions, and the data confirmed by at least
two individuals. Only two of these cDNA were identical with those of
previously reported boIFN- (denoted as 1a and 1b in Fig. 1 and Table 1
). The remaining 22 cDNA
were novel and theoretically encoded 17 different proteins. However,
only twelve of these cDNA could be assumed to represent unique
polymorphic forms on the criterion that the nucleotide substitutions
were noted in more than one PCR. These twelve cDNA encoded six distinct
proteins (1c, 2a, 2b, 3a, 3b, and 3e; Fig. 1 and
Table 1).
|
2c, AF196323; 3c, AF196326; and 3d, AF196327). Each
contains a single nonsynonymous base change that distinguishes it from
any other boIFN-.
The inferred amino sequences of both the previously cloned (1a,
1b, and 1d) and newly identified boIFN- (minus their signal
sequences, all of which were identical) are aligned in Fig. 1. The
differences in amino acid composition are summarized in Table 1. A
number of cDNA that differed from each other in base sequence but that
did not provide a change in amino acid residues are not included in
Fig. 1, but are included in Table 1. For example, five distinct cDNA
sequences encoded the same protein, 1c (Table 1).
When all the confirmed sequences were compared, they fell into a
minimum of three groups based on phylogenetic analysis of nucleotide
(Fig. 2B) and amino acid differences
(Fig. 2B). Although bootstrapping indicated that the trees were not
statistically robust, particularly in some of their outer branches, the
two approaches gave identical branching patterns (Fig. 2, A and B).
|
). None of the cDNA differed by more
than 1.8% in base sequence from the original boIFN-1 clone isolated
by Imakawa et al. (12), which is termed
boIFN-1a throughout this report. Table 1 provides a classification
scheme for the boIFN- protein based on amino acid difference and the
phylogenetic analysis from Fig. 2. It includes all the previously
identified boIFN- as well as the novel cDNA identified from the
present work. The range of sequence differences was somewhat greater
for amino acids than for nucleotides, with boIFN-1a differing from
its most distant relatives, 3b, 3c, and 3d, by 3.5% (Table 2).
|
forms in bovine blastocysts transcripts expressed in
bovine blastocysts was conducted in a series of further RT-PCR
amplification experiment on RNA extracted from bovine blastocysts
obtained by in vitro maturation/in vitro
fertilization/in vitro culture procedures. This study also
allowed further verification of sequences already determined and
provided a crude measure of relative transcript abundance within the
pooled RNA. A total of 70 cDNA were subjected to complete
bi-directional sequencing. Fifty-nine of these sequences corresponded
to ones listed in Table 1. Eleven were unique, differing by no more
than three bases from other cloned cDNA. Because it is again unclear
whether or not these unique sequences are the outcome of errors during
PCR amplification, they have been temporarily laid aside and not
entered into GenBank. Curiously, of the 59 confirmed sequences, 26
corresponded to boIFN-1c and 26 to boIFN-3a. BoIFN-1b, -3b,
and -1a were represented by three, three, and one cloned cDNA,
respectively. There were no cDNA representing 2a, 2b, 2c,
3b, 3c, and 3d. This skewed pattern was seen in the day7
blastocysts (ten boIFN-1c, nine 3a and one 1b), the day-9
blastocysts (seven boIFN-1c, thirteen 3a, and one 3e) and in
the blastocyst outgrowths (nine boIFN-1c, four 3a, one 1a,
one 1b, one 3e).
Ribonuclease protection assays
It was of interest to determine whether boIFNT from the
three main phylogenetic groupings showed similar patterns of expression
over development. Therefore, ribonuclease-protection assays were used
to quantify differences in mRNA abundance of three IFNT,
with one representative from each group (Fig. 3, A and B). Three riboprobes were chosen
(ones for boIFNT-1a, -2b, and -3d). The relative abundance
of mRNA at day 14 and day 25, when expression was anticipated to be
low, and at day 17 and day
19 when boIFN- production is considered to be
close to maximal (22, 24, 32). Each mRNA was detected at
all four stages, although expression was very low in the day
25 material (Fig. 3A). As only single pools of
RNA were available at day 14 and day
25, no SE term could be
calculated to provide between animal or between pool variation (Fig. 3B). However, the day 14 material was obtained
from many conceptuses, the day 25 from six.
Therefore, the mean values for these days are likely to be accurate,
even though the experimental error between assays on the same samples
of RNA was quite high because the protected bands were of relatively
low intensity relative to background. (Data not shown.) The day
17 and day 19 RNA were from
two different pools, so that calculation of standard errors was
possible. There was no difference in the concentration of IFN- mRNA
on these two days. Mean concentrations of mRNA summed for all three
forms differed over time with values at day 17
higher (P < 0.05) than at other stages of pregnancy
(Fig. 3B). The amount at day 19 showed a tendency
to be higher than values at day 14 and day
25.
|
). As these fragments were not detected in the undigested probe
preparations, they do not represent protection by partially transcribed
probes. It seems more likely that these extra bands reflect the
presence of mRNA that were similar to, but not identical in sequence to
those being studied. The abundance of some of these fragments was
comparable to that of the full-length fragments after their
densitometric units were adjusted for their fragment size (quantitative
data not shown).
A statistical analysis of the main effect, that of all three -mRNA
forms over different days, revealed that the concentration of any one
mRNA did not differ significantly from those of the two others on the
four days analyzed (P > 0.1). These data strongly
suggest that the three genes contributed equivalent amounts of mRNA
throughout the peri-implantation period of development.
Sequence analysis of IFN- genes (IFNT)
To determine whether IFNT from different phylogenetic
groupings had conserved or very different promoter sequences,
particularly within the main enhancer regions, an extensive series of
IFNT clones were analyzed after PCR amplification from
genomic DNA. The primers were designed to provide approximately 430 bp
of DNA upstream of the transcription start site, the 5' untranslated
region and the entire open reading frame. This upstream region is well
conserved across ruminant ungulates (19) and is both
necessary and sufficient to provide full expression from IFNT promoters
in JAr choriocarcinoma cells (33, 34). One pass sequencing of the
coding regions of 12 clones indicated that the genes fell into the
1, 2, and 3 groupings noted for the cDNA. Four of the twelve
were bi-directionally sequenced (Fig. 4).
Two of them (T1A and T1C) were identical in their
coding regions to the cDNA for IFN-1a and IFN-1c, respectively
(data not shown). The third (T2B) and fourth
(T3B) clones differed at a single nucleotide within their
promoter regions (Fig. 4), but encoded different proteins (IFN-2b
and IFN-3b, respectively). The promoter regions of these four genes
are compared with that of the IFNT1A gene (GenBank M60903)
in Fig. 4. All five sequences were very similar, and that of
T1A was identical with that of the previously reported
sequence for boIFNT1A (14). The Ets-2/AP1
enhancer region (-77 to -69) (35) was completely
conserved in all four genes analyzed.
|
protein production and activity proteins were generated per liter of fermentation medium. All
four boIFN- had a molecular mass (19 to 20 kDa), which would be
expected for unglycosylated IFN- (Fig. 5A). All recombinant preparations
cross-reacted with antiserum generated against boIFN-1a (Fig. 5B).
This antiserum was specific for IFN- because it did not cross-react
with boIFN-1 (Ciba-Geigy Ltd., Basel, Switzerland) or
Escherichia coli extracts (Fig. 5B).
|
forms had the ability to protect MDBK and
BEND cells from lysis following vesicular stomatitis virus infection
(Table 3). Forms 1a, 1c, and 2b
were equivalently potent. The 3b protein had significantly reduced
activity on both cell lines when compared with the other three
proteins. Although closely similar in sequence, the different boIFN-
did not have identical biological potencies as defined by their
antiviral activities on two bovine cell lines.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
transcripts. At present there are
12 novel sequences that we are confident represent different boIFN-
sequences and an additional 21 cDNA (including those in the second
series of amplifications from blastocyst RNA) whose sequence remain
questionable as to whether they represent additional polymorphic
sequences or are simply byproducts of RT-PCR-induced base mutations.
The base substitution rate (1 in 1092 nucleotides), arising as a
consequence of RT and 30 rounds of PCR performed under relatively
standard conditions, was within the range reported by others
(36, 37, 38, 39) and is about that expected from supplier’s
specifications for the proof reading polymerase
(1x10-6 mutations/bp/duplication). Such multiplicity is also found among other type 1 IFN, such as the IFN-, but the latter are involved in
protecting host cells from viral pathogens and might be expected to be
evolving at a high rate (1). Why the IFNT might
be duplicating so frequently is discussed later.
On the basis of phylogenetic analysis (Fig. 2 and Table 1), the
sequences fall into three main groups. Group 3 appears to be the most
ancient, but overall similarities are high. The average base
substitution rate within the IFNT of the pecoran ruminants
has been calculated by the Kimura two-parameter method
(40) to be approximately 0.27 per 100 bases per million
years (1). On this basis, the divergence time between
group 3 and the lineage leading to groups 1 and 2 (Fig. 2A) is between
6 and 7 million years. It is generally accepted that the ancestors of
the Bovinae (cattle and their relatives) and Caprinae (sheep and goats)
diverged about 20 million years ago (41). This time is,
however, long compared with the years since first domestication of
cattle >10,000 yr ago (42) and particularly during the
short period (
200 yr) that selection was being used to develop the
modern dairy and beef breeds of Bos taurus. Therefore, even
though the expressed boIFNT represent a rapidly evolving
group of genes, it is likely that the majority of them are represented
in all Bos taurus breeds. Their uniformity contrasts with
the diversity of the ovIFNT, which differ as much as 10% in
nucleotide sequence (28). It will be of interest to
determine whether, in cattle, a more ancient set of pseudogenes exists
that can form the link to the IFNT of sheep.
Analysis of the nucleotide changes reveals another interesting feature
of the boIFNT cDNA. When pairwise comparisons were made
across the entire group of nine sequences (40), it becomes
evident that the number of base substitutions at nonsynonymous sites,
i.e. amino acid changing, is only slightly less than the
substitution rate at synonymous (neutral) sites (1.2 bases/100 sites
vs. 1.7 bases/100 synonymous sites). Because there are more
of the former, amino acid sequences are relatively more divergent than
nucleotide sequences. In general, amino acid changes are
disadvantageous and are removed by purifying selection
(43). With the boIFN-, selection appears to be favoring
amino acid change (43). As Haig (44) has
pointed out, the interface between the placenta and endometrium is
likely a site where conflict between maternal genes is most intense and
where rapid rates of genetic change are most expected. This process may
lead to rapid, adaptive diversification of the IFNT.
It is important to note that each of the expressed boIFN- retains
the single potential site for N-glycosylation at Asn78. The observation
is consistent with earlier data that all forms of boIFN- are
glycosylated (22, 23, 24). The lack of glycosylation of the
recombinant forms (Fig. 5 and Table 3) could provide products that are
less active and possibly less stable than the forms with carbohydrate,
as has been observed for recombinant human IFN-ß (45).
Substitution of Gly105 with Glu, Gly126 with Asp, and Glu135 with Val
clearly provide the potential charge differences that could account for
the multiply charged forms of native boIFN- noted during isoelectric
focusing (22, 23, 24). In addition, these changes in charge
correspond with the separation of boIFN- forms into their three
classes, and therefore may serve as markers for each class.
Nuclease protection was used to estimate the relative amounts of
different classes of transcript in pooled samples of RNA during the
period of boIFN- expression. (Fig. 3, A and B). The data confirm
observations of others (22, 24, 32) that the zenith of
boIFN- production occurs around day 17. They also
indicate that different genes, or at least different classes of genes,
are most likely transcribed at comparable rates during the period
between day 14 and day 25. It is of interest,
therefore, why the amplification of RNA from blastocysts showed such
bias toward just two forms, namely boIFN-1c and boIFN-3a, with
some forms not being represented at all. As the RNA was collected from
over 200 embryos sired by six bulls, it seems unlikely that the bias
resulted from the narrow genetic base of the blastocysts analyzed.
Experimental bias was also expected to be minimal, as all of the
transcripts were of the same length, varied less than 2% in nucleotide
sequence, and were amplified by using primers representing completely
conserved regions of the genes. As the blastocyst is the first
developmental stage at which IFN- is expressed (25, 46), one conclusion is that there is preferential transcription
from two of the nine genes in blastocysts during the early onset phase
of expression.
Analysis of the promoter regions from different IFN- genes that had
been amplified from genomic DNA revealed that the sequences again fell
into the same three groupings used to classify the cDNA and proteins.
The promoter regions differed at only a few positions over 430 bp
upstream of the transcription start site. There were no differences
within the Ets2/AP1 enhancer site (-77 to -69; Fig. 5), which is
mutated in certain ovIFN- genes that appear not to be expressed
(35). This high degree of conservation probably explains
why different bovine genes are expressed similarly in day
14 to day 25 conceptuses.
In conclusion, several forms of boIFN- are expressed by the
developing bovine conceptus during the day 14 to day
25 period, during which time the trophoblast elongates,
intervenes to prevent corpus luteum regression, and begins
placentation. Relative transcript abundance was similar for several of
these forms throughout this period, indicating that the genes are
likely to be under similar transcriptional control. It remains unclear,
however, why there is a need for such a variety of IFNT. One
possibility is that the simultaneous expression of several
IFNT fulfills a need for large quantities of IFN- protein
(2, 3, 5). Another is that there is strong selective
pressure for emergence of IFN- proteins that better represent the
interests of the fetus in its interaction with the mother.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current addresses: Dr. Alan D. Ealy, Department of Dairy and Animal
Science, Pennsylvania State University, 304 Henning Building,
University Park, Pennsylvania 16802; Dr. Sandra F. Larson, Department
of Biology, Furman University, Greenville, South Carolina 29613; Dr.
Limin Liu, Division of Pulmonary Medicine, Department of Medicine, 317
Medical Science Research Building, Box 2612 Medical Center, Duke
University, Durham, North Carolina 27710; Dr. H. Michael Kubisch,
Tulane Regional Primate Research Center, 18703 Three Rivers Road,
Covington, Louisiana 70433.
3 These authors contributed equally to the work and should be
regarded as co-first authors of the paper.
4 Received a postdoctoral fellowship from the Food for the
Twenty-First Century Program at the University of Missouri.
5 Received salary support from the Howard Hughes Undergraduate
Research Internship Program.
Received January 24, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
regulates expression of
endometrial receptors for estrogen and oxytocin but not progesterone.
Biol Reprod 53:732–745[Abstract]
modulates phorbol
ester-induced production of prostaglandin and expression of
cyclooxygenase-2 and phospholipase-A2 from bovine
endometrial cells. Biol Reprod 63:417–424II possess distinct 5'-promoter and 3'-flanking
sequences. J Biol Chem 266:3060–3067 genes encoding functional class II and trophoblast
(oTP-1) -interferons. Mol Cell Endocrinol 76:161–171[CrossRef][Medline]
genes are not expressed
identically and their protein products display different activities.
Biol Reprod 58:566–573This article has been cited by other articles:
 |
|
 |
|
  F. N T Cooke, K. A Pennington, Q. Yang, and A. D Ealy Several fibroblast growth factors are expressed during pre-attachment bovine conceptus development and regulate interferon-tau expression from trophectoderm Reproduction, February 1, 2009; 137(2): 259 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ezashi, P. Das, R. Gupta, A. Walker, and R. M. Roberts The Role of Homeobox Protein Distal-Less 3 and Its Interaction with ETS2 in Regulating Bovine Interferon-Tau Gene Expression-Synergistic Transcriptional Activation with ETS2 Biol Reprod, July 1, 2008; 79(1): 115 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. D. Michael, I. M. Alvarez, O. M. Ocon, A. M. Powell, N. C. Talbot, S. E. Johnson, and A. D. Ealy Fibroblast Growth Factor-2 Is Expressed by the Bovine Uterus and Stimulates Interferon-{tau} Production in Bovine Trophectoderm Endocrinology, July 1, 2006; 147(7): 3571 - 3579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ezashi and R. M. Roberts Regulation of Interferon-{tau} (IFN-{tau}) Gene Promoters by Growth Factors that Target the Ets-2 Composite Enhancer: A Possible Model for Maternal Control of IFN-{tau} Production by the Conceptus during Early Pregnancy Endocrinology, October 1, 2004; 145(10): 4452 - 4460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Glover and G. E. Seidel Jr Increased Messenger RNA for Allograft Inflammatory Factor-1, LERK-5, and a Novel Gene in 17.5-Day Relative to 15.5-Day Bovine Embryos Biol Reprod, September 1, 2003; 69(3): 1002 - 1012. [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]
|
|
|
|
 |
|
 M. A. Larson, K. Kimura, H. M. Kubisch, and R. M. Roberts Sexual dimorphism among bovine embryos in their ability to make the transition to expanded blastocyst and in the expression of the signaling molecule IFN-tau PNAS, July 24, 2001; (2001) 171305398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Larson, K. Kimura, H. M. Kubisch, and R. M. Roberts Sexual dimorphism among bovine embryos in their ability to make the transition to expanded blastocyst and in the expression of the signaling molecule IFN-tau PNAS, August 14, 2001; 98(17): 9677 - 9682. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
