0013-7227/99/$03.00/0
Endocrinology
Copyright © 1999 by The Endocrine Society
Vol. 140, No. 5
Printed in U.S.A.
Suppression Subtractive Hybridization Identified a
Marked Increase in Thrombospondin-1 Associated with
Parturition in Pregnant Sheep Myometrium*
WEN XUAN WU, QI ZHANG, XIAO HONG MA, NOBUYA UNNO,
PETER W. NATHANIELSZ
AND
Laboratory for Pregnancy and Newborn Research (W.X.W., Q.Z., X.M., P.W.N.), Department
Physiology, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853-6401; and
Department of Obstetrics and Gynecology (N.U.), Faculty of Medicine, University of Tokyo, 113-0033
Tokyo, Japan
ABSTRACT
Changes in several extracellular matrix proteins, such as fibronectin in fetal membrane and cervical collagens, occur at term and preterm delivery. However, no studies have evaluated changes in extracellular matrix proteins, in relation to myometrial activation recorded
using invasive techniques during parturition in any species. We used
suppression subtractive hybridization (SSH), to demonstrate the dramatic increases in one extracellular matrix protein, thrombospondin-1 (TSP1), in the pregnant ovine myometrium associated
with parturition. Myometrial poly-A1 RNA, extracted from term control ewes not in labor at 143–147 days of gestational age (dGA, n 5
4), and from ewes in spontaneous term labor (STL) at 145–147 dGA
(n 5 4), was subjected to SSH to construct a subtracted myometrial
complementary DNA library. A complementary DNA clone from myometrial subtracted library, representing differentially expressed gene
in the pregnant sheep myometrium during STL, was identified as
TSP1 by sequence analysis and Blastn search. This cloned TSP1 was
used to perform Northern blot analysis on total RNA isolated from five
early controls, not in labor, at 130 dGA, five pregnant ewes in betamethasone-induced premature labor (BPL) at 130 dGA (betamethasone administered iv to the fetus at 0.48 mg over 48 h), six term-
P
REMATURE birth is a major problem of pregnancy,
accounting for 75% of perinatal mortality and 50% of
long-term morbidity (1). The development of therapeutic
regimens that will effectively prevent preterm birth awaits a
greater understanding of the mechanisms that regulate labor.
A more profound understanding of the parturition requires
integration of processes that maintain pregnancy and inhibit
parturition, as well as mechanisms that activate and stimulate myometrial contraction.
The switch in myometrial contractility patterns, from contractures to contractions (2, 3), is accompanied by myometrial
activation (1) involving changes in expression of a collection
of contraction-associated proteins. Contraction-associated
proteins that have been investigated during late gestation
and labor in sheep and other species include: the estrogen
receptor (4, 5); the steroid receptor-associated proteins-heat
Received May 19, 1998.
Address all correspondence and requests for reprints to: Dr. Peter W.
Nathanielsz, Laboratory for Pregnancy and Newborn Research, Department Physiology, College of Veterinary Medicine, Cornell University,
Ithaca, New York 14853-6401. E-mail: [email protected].
* This work was supported by Grant NIH-HD-21350.
control-not-in-labor ewes at 143–147 dGA, and six pregnant ewes in
STL. Northern blot analysis demonstrated that TSP1 increased significantly, associated with both BPL and STL. TSP1 protein level
paralleled the increase of TSP1 messenger RNA during BPL and STL.
To determine whether increase in this gene paralleled the increased
strength of myometrial contractility, six ewes were treated with nimesulide, a selective PG synthase 2 inhibitor, at 147–148 dGA. Nimesulide infusion to the ewe iv to inhibit myometrial contraction (30 mg
bolus, followed by 5 h infusion, 30 mg/h) commenced 9 h after onset
of labor at 147–148 dGA. TSP1 in the myometrium decreased when
myometrial contraction was inhibited by nimesulide. Both in situ
hybridization and immunocytochemistry demonstrated that fibroblasts and the smooth muscle cells contained TSP1 messenger RNA
and protein. TSP1 was also localized in the extracellular matrix. Our
conclusions are: 1) our data provide the first evidence that changes in
TSP1 are associated with myometrial activation in pregnant sheep
during term and preterm labor; 2) myometrial fibroblasts and the
smooth muscle cells are responsible for producing TSP1; and 3) SSH
is a powerful technique that enables us to study differentially regulated genes during labor. (Endocrinology 140: 2364 –2371, 1999)
shock proteins 70 and 90 (Hsp 70 and 90) (6); oxytocin (OT)
(7–9); agonist receptors—OT receptors (10 –13), PG receptors
(14 –16), gap junction proteins (17, 18), and ion channels (19,
20). Once activated, the myometrium can respond to paracrine and endocrine stimulation by uterotonic agonists (e.g.
OT and PGs) that bring labor to a successful conclusion.
Based on these observations, others and we (1, 21) have
proposed that the onset of labor is associated with preparation of the myometrium, with resultant changes in a cassette
of genes (up- or down-regulation). Each member of the cassette of genes contributes, to some extent, in the switch in
myometrial contractility patterns from contractures to contractions, as well as the progression of established labor.
Altered abundance of members of this cassette of genes in the
myometrium is a prerequisite for labor and delivery. Myometrial activation before and during labor seems to play a
critical role in connecting fetal signals that indicate fetal
readiness for birth, to maternal factors that carry labor forward to expel the fetus.
Understanding of the molecular basis of myometrial activation is incomplete. More information is needed on the
genes involved during labor and the factors regulating their
2364
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
expression. To understand the molecular basis of myometrial
activation, the relevant subsets of differentially expressed
genes associated with the various stages of parturition must
be identified, cloned, and studied in detail. Suppression subtractive hybridization (SSH) is a powerful approach to identify and isolate complementary DNAs (cDNAs) of differentially expressed genes. In the present study, we employed
this sensitive and powerful technique to identify unknown
factor(s), which may play important roles in the process of
parturition. Improved understanding of the cellular processes that occur at labor will advance our knowledge of the
key regulatory mechanisms involved in preparation for and
development of parturition.
Using SSH, we identified a marked increase at the messenger RNA (mRNA) level associated with parturition in one
extracellular matrix protein, thrombospondin-1 (TSP1, a glycoprotein) in the pregnant ovine myometrium. The increase
in TSP1 mRNA level was accompanied by increased TSP1
protein level, measured by Western blot analysis. We also
examined the cellular distribution of TSP1 mRNA, in relation
to its protein expression in the pregnant sheep myometrium,
using in situ hybridization and immunocytochemistry.
Materials and Methods
Animals and tissue collection
Twenty-eight pregnant Rambouillet-Dorset ewes from the University
flock, bred on a single occasion and carrying fetuses of known gestational age, were studied. Experimental procedures were approved by
the Cornell University Institutional Animal Care and Use Committee.
The Cornell facilities are approved by the American Association for the
Accreditation of Laboratory Animal Care. At 120 days of gestational age
(dGA), ewes from which tissues were obtained were instrumented with
electromyogram (EMG) leads, sewn into the myometrium, and fetal and
maternal carotid arterial and jugular venous catheters (2). Labor was
defined as having occurred when the myometrial EMG record showed
a clear switch from contractures to contractions, followed by contraction
activity for at least 5 h (22).
Myometrium was obtained from ewes, during labor, in a comparative
study to determine whether there were any differences in labor induced
by the infusion of betamethasone [Celestone phosphate, ScheringPlough Corp. (Kenilworth, NJ), n 5 5]. Betamethasone (10 mg/h) was
administered iv into the fetal jugular vein continuously over a period of
48 h, in a total dosage of 0.48 mg, to precipitate betamethasone-induced
premature labor (BPL). Ewes underwent necropsy at 130 dGA, when
they went into labor as a result of the betamethasone administration.
Myometrium was also obtained from contemporary control ewes (n 5
5) at the same stage of gestation (130 dGA). These ewes were designated
as early controls, not in labor (ECNL). Fetuses of ECNL ewes were
infused with physiological saline. These ewes were not in labor, because
the myometrial EMG showed only contractures and no contractions.
Tissues were also collected from ewes in spontaneous term labor (STL)
at 145–147 dGA (n 5 6), and term control ewes not in labor (TCNL; n 5
6) at the same gestational age (143–147 dGA).
Nimesulide infusion. Nimesulide (D-5648; Sigma Chemical Co., St. Louis,
MO) was prepared in DMEM at a concentration of 1.25 mg/ml. Nimesulide infusion to the ewe (n 5 6) iv (30 mg bolus, followed by 6 h
infusion, 30 mg/h) commenced 9 h after onset of labor, at 147–148 dGA.
Myometrium was always removed from the ventral aspect of the
midportion of the body of the uterus and frozen in liquid nitrogen for
later RNA extraction. Another similar portion of the myometrium was
embedded in Tissue Freezing Medium (Triangle Biomedical Science,
Durham, NC) and frozen in liquid-nitrogen-cooled isopantane for later
study by in situ hybridization analysis and immunocytochemistry. Frozen tissues were stored at 280 C.
2365
Construction of subtracted cDNA library
Total RNA and poly-A1 RNA isolation. Total RNA was isolated from
myometrium of the sheep in STL and TCNL, as described previously (4).
Briefly, total RNA was isolated from frozen tissues by homogenization
in 4.2 m guanidinium thiocyanate solution. RNA was pelted through a
5.7 m cesium chloride cushion. The RNA purity and recovery of each
tissue was determined by UV spectrophotometry (260 and 280 nm).
Poly-A1 RNA was extracted from total RNA using the MicroFastTrack
kit as suggested by the manufacturer (Invitrogen, San Diego, CA). The
poly-A1 RNA from the two groups (TCNL and STL) was purified in
parallel using the same reagents and protocol.
cDNA subtraction library. SSH (23) was used to construct subtraction
library to identify up-expressed genes in the STL myometrium. The
library was made using PCR-Select cDNA Subtraction Kit, according to
the protocol provided by the manufacturer (CLONTECH Laboratories,
Inc., Palo Alto, CA). Briefly, the SSH protocol is as follows: poly-A1 RNA
from STL and TCNL were designated as tester and driver, respectively.
The double-strand cDNAs were synthesized from 2 mg myometrial
poly-A1 RNA obtained from both STL and TCNL ewes. The synthesized
double-strand cDNAs were then digested with restriction enzyme RsaI.
The digested blunt ends of the tester cDNA (from STL) were divided into
two parts and ligated with two different cDNA adaptors. Two cycles of
hybridization were followed after the ligation. In the first cycle, excess
driver cDNAs (from TCNL) were added to the two populations of tester
cDNAs after denaturation, to achieve equalization and enrichment of
differentially expressed sequences in the hybridization solution. During
the second hybridization process, denatured fresh driver cDNAs were
added into the two tester cDNA populations again and allowed to
anneal, to further enrich the differentially expressed sequences. The final
hybridization solution (also called the subtracted library) was employed
as a template to amplify the differentially expressed sequences in the
tester population by using a set of PCR primers and was followed by
nested PCR primers. Both sets of PCR primers were annealed to the
different sites of ligated adaptors. The secondary PCR amplification, by
using nested primers, considerably enriched the differentially expressed
sequences and reduced background of PCR product (23).
The amplified cDNA fragments from the secondary PCR were ligated
into a pCR2.1 vector using a T/A cloning kit (Invitrogen). Positive clones
were identified by differentially screening the subtracted library using
both forward-subtracted and reverse-subtracted probes (derived by
switching between tester and driver populations for another subtractive
hybridization). The forward-subtracted and reverse-subtracted probes,
labeled with a-32P deoxycycidine triphosphate (dCTP), were hybridized
with the replicate cDNA library dot blots (Gene Screen Plus membrane,
NEN Research Products, DuPont NEN, Boston, MA). Positive clones,
identified by differential screening, were cultured in LB medium to
obtain plasmid DNA for sequencing. The plasmid DNA was purified
using the Biorobot system (Qiagen, Chatsworth, CA), and sequencing
was done by Taq cycle sequencing using DyeDeoxy terminators in an
automated sequencer (ABI 377 DNA sequencer; PE Applied Biosystems,
Foster City, CA) at the core sequencing facility of Cornell University. The
cloned fragments were sequenced from at least two different clones and
from both the coding and noncoding strands. The acquired sequence
data were aligned against the GenBank nucleotide database at the National Center for Biotechnology Information (National Institutes of
Health, Bethesda, MD) using Blastn to search for sequence matches.
Identification of TSP1 as an up-regulated gene in parturition. The Blastn
search revealed that three of the up-regulated gene clones were homologous to ovine TSP1. Because it was isolated from the tester cDNA
population (STL myometrium), it represented the up-regulated gene
from the myometrium of the pregnant sheep during labor. The acquired
sequence data were aligned against the GenBank nucleotide database
and revealed that the clones contain partial coding and most downstream sequences of TSP1 mRNA. The cloned ovine TSP1 fragment was
then used as a cDNA probe for Northern blot analysis and in situ
hybridization, as described below.
Northern blot analysis
Total RNA was prepared from individual tissues, as described above.
Samples of total RNA (30 mg/lane) from each tissue were denatured in
2366
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
17.4% (vol/vol) formaldehyde, 50% (vol/vol) formamide, 20 mm 3-(Nmorpholino) propanesulfonic acid), 5 mm sodium acetate, and 1 mm
EDTA (pH 7.0) for 5 min at 65 C and separated on a 1% (wt/vol)
agarose/0.66 m formaldehyde gel. Ethidium bromide-stained ribosomal
RNA (rRNA) bands were visualized (UV) to ensure that RNA degradation had not occurred and that an equal amount of RNA had been
loaded into each lane. After electrophoresis, RNA was transferred to a
nylon membrane (Gene Screen Plus; DuPont NEN, Wilmington, DE) by
capillary blotting for 24 h in 10 3 SSC (1 3 SSC is 0.15 m NaCl and 0.015
m Na Citrate, pH 7.0). Completion and uniformity of transfer were
assessed by determining transfer of 28S and 18S rRNA from the gel.
Membranes were prehybridized at 42 C for 5 h in hybridization solution
[50% (vol/vol) deionized formamide, 50 mm sodium phosphate, 0.8 m
NaCl, 2% (wt/vol) SDS, 100 mg salmon sperm DNA/ml, 20 mg transfer
RNA/ml, and 1 3 Denhard’s (503 5 1% solution of BSA, Ficoll, and
polyvinylpyrrolidone).
Endo • 1999
Vol 140 • No 5
The cloned ovine TSP1 cDNA fragment was labeled with a-32P dCTP
(for Northern blot analysis) and labeled with a-35S dCTP (for in situ
hybridization) using the random priming method (DuPont NEN) to
specific activities of approximately 1 3 109 cpm/mg and used at a final
concentration of 1 3 106 cpm specific probe/ml of hybridization
solution.
Hybridization was carried out at 42 C for 20 h in hybridization
solution containing the specific TSP1 cDNA probe. Membranes were
washed sequentially in 2 3 SSC at room temperature for 10 min, and
0.5 3 SSC with 0.1% SDS at 65 C for 30 min. Kodak X-OMAT AR film
(obtained from Sigma Chemical Co.) was exposed to the membrane with
intensifying screens at 280 C. Exposure durations were varied to achieve
hybridization signals within the limited linear range for densitometry.
Membranes were stripped of TSP1 probe by boiling in 0.1 3 SSC with
0.1% (wt/vol) SDS for 30 min and rehybridized with a-32P dCTP-labeled
18S cDNA probe to normalize TSP1 mRNA levels. Autoradiographed
signals were quantified by Scan densitometry.
In situ hybridization
FIG. 1. Acrylamide gel analysis of the secondary PCR products of
subtracted and unsubtracted ovine STL myometrial samples. Lane 1,
Subtracted STL myometrial samples; lane 2, unsubtrated STL myometrial samples; bp, indicated DNA molecular weight (M) in bp. A
1.2-kb band, indicated by the white arrow, and other bands with
different sizes represent the differentially expressed genes during
labor in the STL myometrial subtracted sample.
In situ hybridization was performed as described previously (13),
with some modification. Frozen sections of pregnant sheep myometrium
were cut and mounted on poly-l-lysine-coated slides. Tissue sections
were fixed in freshly prepared 4% paraformaldehyde in 10 mm PBS (pH
7.2) for 10 min, followed by 2 3 PBS wash, 5 min each, at room temperature. Prehybridization was performed at 42 C for 1 h in prehybridization buffer containing 50% formamide, 5 3 SSPE (1 3 SSPE being 0.18
m NaCl, 10 mm NaH2PO4, and 1 mm EDTA), 0.1% SDS, 0.1% (vol/vol)
Denhardt’s solution, 200 mg denatured salmon testis DNA/ml, and 200
mg transfer RNA/ml. Hybridization was carried out for 18 h at 42 C in
hybridization buffer [prehybridization buffer plus 4% (wt/vol) dextran
sulfate] containing 1.0 3 106 cpm 35S-labeled ovine TSP1 cDNA probe/
section. After hybridization, the sections were rinsed at room temperature for 2 h in 2 3 SSC, 2 h in 1 3 SSC, 1 h in 0.5 3 SSC, and finally
for 1 h in 0.5 3 SSC at 37 C. The sections were then dehydrated by passing
through an alcohol series, containing 300 mm ammonium acetate, and
coated with liquid photographic emulsion (NTB2, Kodak). After 10 days
exposure, the sections were developed and stained with hematoxylin
and eosin.
FIG. 2. Dot blot analysis of differential screened up-expressed gene clones after STL myometrial SSH. Replicate dot blots were hybridized with
both forward-subtracted (right panel) and reverse-subtracted (left panel) probes. Dots hybridized with reverse-subtracted probes correspond
to background (left panel). Clones hybridized only with forward-subtracted probes represent mRNAs that are differentially expressed in the
tester (STL) population of myometrium (right panel). All of the clones in red circles in the right panel are the up-regulated genes in the STL
myometrial samples. Three of these clones, indicated by black arrows, were identified as TSP1, by sequence analysis and Blastn search.
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
FIG. 3. Northern blot analysis of TSP1 mRNA (A) in the pregnant
sheep myometrium in BPL (lanes 6 –10) and STL (lanes 16 –20),
compared with ECNL (lanes 1–5) and TCNL (lanes 11–15). B, The
same blot was probed for 18S; C, densitometric analysis of the ratio
of TSP1 mRNA and 18S rRNA in myometrium during both BPL and
STL, compared with ECNL and TCNL (n 5 5 for each group). Significant increases were found in both groups of ewes in labor, when
compared with ECNL and TCNL (mean 6 SEM). *, P , 0.01, BPL and
STL vs. ECNL and TCNL.
Controls
Serial sections were treated with pancreatic ribonuclease (RNase) A
(20 mg/ml), for 30 min at room temperature, before hybridization. After
enzyme treatment, the sections were rinsed in three changes of 2 3 SSC
(5 min each) and hybridized with the a-35S dCTP-labeled ovine TSP1
cDNA probe, as described above.
Western blot analysis
Cytosolic extracts. Myometrium was ground into small pieces and then
homogenized in Tris buffer [50 mm Tris (pH 7.4), 10 mm EDTA, and 1
mm diethyldithi-ocarbamic acid], containing 2 mm octyl glucoside, and
was centrifuged at 30,000 3 g for 20 min at 4 C. The 30,000 3 g supernatant (cytosol) was retained and stored at 280 C until used for electrophoretic analyses. The protein concentration was determined by the
method of Bradford (Bio-Rad Laboratories, Inc., Hercules, CA).
The cytosolic proteins (50 mg/lane) were then separated by 10%
SDS-PAGE and were electrophoretically transferred to a nylon membrane (Imobilon, Millipore Corp., Bedford, MA) using a transfer blot cell
(Bio-Rad Laboratories, Inc.). The filters for immunostaining were
blocked with 2% BSA in 10 mm Tris-Cl buffer containing 0.1% Tween-20.
After blocking, the blots were washed three times with wash buffer (5
min each), containing 10 mm Tris-Cl and 0.1% Tween-20, and were
incubated with a mouse monoclonal anti-TSP1 antibody (NeoMarkers,
Fremont, CA; Cat No: MS-421-P1, clone A6.1, 1:400 dilution) at 4 C
overnight. The blots were then washed, incubated with peroxidase-
2367
FIG. 4. Northern blot analysis of TSP1 mRNA (A) in the pregnant
sheep myometrium in TCNL (lanes 1– 6), STL (lanes 7–12), and Nimesulide-infused ewes (Nim, lanes 13–18). B, The same blot was probed
for 18S; C, densitometric analysis of the ratio of TSP1 mRNA and 18S
rRNA in myometrium from TCNL (n 5 6); STL (n 5 6), and Niminfused ewes (n 5 6). There was a significant increase in TSP1 mRNA
during STL (**, P , 0.01). After NIM infusion, TSP1 mRNA decreased
significantly in myometrium (*, P , 0.05; mean 6 SEM).
conjugated sheep antimouse IgG (Amersham Life Sciences, Arlington
Heights, IL) at room temperature for 1 h. After each antibody incubation,
the blots were washed three times (15 min each) in wash buffer. The
protein bands were visualized using an enhanced chemiluminescence
Western blotting detection kit (Amersham). The molecular size of the
proteins was determined by running standard molecular weight marker
proteins (Bio-Rad Laboratories, Inc.) in an adjacent lane.
Immunocytochemistry
Frozen sections (4 mm) of ovine myometrial tissue were immunostained for TSP1 using the avidin-biotin immunoperoxidase method.
Briefly, sections were fixed in acetone for 10 min at room temperature
and then rinsed twice in 0.05 m Tris buffered saline, 5 min each. Unless
otherwise specified, all slides were sequentially incubated, for various
times, at room temperature, with each of the following reagents: 1) 3%
(vol/vol) H2O2 in methanol for 30 min to block endogenous peroxidase
activity; 2) 25% (vol/vol) normal horse serum and 5% (wt/vol) BSA in
0.05 m Tris-Cl with 0.15 m NaCl, pH 7.6, for 1 h; 3) mouse monoclonal
anti-TSP1 antibody (1:400 dilution) described above as used for Western
blotting, for 20 h at 4 C; 4) biotinylated horse antimouse IgG (Vector
Laboratories, Inc., Burlingame, CA) for 0.5 h; 5) avidin-biotin complex
(Vector Laboratories, Inc.): 40 ml of each in 5 ml 0.05 m Tris-Cl, pH 7.6,
for 0.5 h; and 6) 3,3-diaminobenzidine tetrahydrochloride (Sigma Chemical Co.) for 10 min: 4 mg/10 ml 0.05 m Tris buffer (pH 7.6), to which was
added 0.2 ml 3% H2O2. After each incubation, the slides were washed
with Tris buffered saline for 15 min, except for step 2). The slides were
then mounted in Permount with or without hematoxylin counterstaining. Specificity of immunostaining for the TSP1 was confirmed by three
approaches: 1) omission of the primary antibody; 2) incubation of the
2368
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
Endo • 1999
Vol 140 • No 5
slides with the normal mouse serum instead of the primary antibody;
and 3) Western blot analysis to determine the specificity of the antibody.
Statistical analysis
After normalization of the content of TSP1 mRNA to 18S rRNA in
individual samples, TSP1 mRNA concentration in each Northern blot
was expressed as a ratio of TSP1 mRNA to 18S rRNA. Differences
between different groups for Northern blot or Western blot analysis
were analyzed by ANOVA followed by multiple comparison using a
Tukey-Kramer procedure. Data throughout are presented as the mean 6
sem.
Results
SSH
Figure 1 demonstrates the electrophoretic analysis of the
secondary PCR product of SSH. There was a different banding pattern between the STL myometrial subtracted sample
(lane 1), compared with the STL myometrial unsubtracted
sample (lane 2). A 1.2-kb band, indicated by the white arrow,
along with other bands of different sizes, can be identified
clearly in the STL myometrial subtracted sample, indicating
a great enrichment of differentially expressed genes in the
subtracted sample. A dot blot of differential screening is
shown in Fig. 2. Duplicate blots were hybridized with forward-subtracted probes and reverse-subtracted probes, respectively. Dots hybridized with reverse-subtracted probes
corresponded to background (Fig. 2). Clones hybridized only
with forward-subtracted probes represent mRNAs that were
differentially expressed in the tester (STL) population of
myometrium (Fig. 2). Three of the clones representing the
TSP1 gene were identified by sequence analysis and Blastn
search.
Northern blot analysis of TSP1
To further confirm the up-regulation of TSP1 mRNA in the
pregnant sheep myometrium during labor, Northern blot
analysis was performed. TSP1 mRNA was significantly increased in ovine myometrial tissue in BPL and in STL (Fig.
3), indicating strong induction of an approximately 6.5-kb
TSP1 mRNA in the myometrium in premature, as well as
term, labor (P , 0.01). After nimesulide infusion, TSP1
mRNA was significantly decreased below levels seen in STL
in the myometrium (Fig. 4) (P , 0.05) but remained significantly higher than TCNL ewes not in labor (P , 0.01).
Western blot analysis of TSP1
To test whether the mRNA changes were carried to the
protein level, Western blot analysis was employed to evaluate TSP1 protein level in ovine myometrium, both before
and during labor. In the pregnant sheep, myometrium immunoreactive TSP1 was expressed in the myometrium with
molecular weights of approximate 120K (Fig. 5). An additional band at 140-kDa was observed in one myometrial
sample. The immunoreactive band of TSP1 increased significantly (P , 0.001) in BPL or STL, which corresponds with the
changes in TSP1 gene transcription observed by Northern
blot analysis.
In situ hybridization and immunocytochemistry
In situ hybridization demonstrated that TSP1 mRNA was
mainly associated with fibroblast-type cells in the myome-
FIG. 5. Western blot analysis of TSP1 protein in the pregnant sheep
myometrium in BPL (A, lanes 6 –10) and STL (B, lanes 6 –10), compared with ECNL (A, lanes 1–5) and TCNL (B, lanes 1–5). C, Mean
relative enzyme protein levels for TSP1 in ovine myometrium from
ECNL, BPL, TCNL, and STL. Values are expressed as mean arbitrary
scale 6 SEM for the enzyme (n 5 5 for each group). Myometrial TSP1
protein level increased significantly (*, P , 0.001), in association with
BPL and STL.
trium (Fig. 6A). However, some of myometrial cells also
contained TSP1 mRNA (Fig. 6B). Control sections treated
with RNase A showed no specific signals (Fig. 6C). Dense
immunostaining for TSP1 was detected in the fibroblast-type
cells, as well as in the extracellular matrix of the pregnant
sheep myometrium (Fig. 6, D1 and D2). Some of the myometrial cells were also stained for TSP1 (Fig. 6E). When
myometrial sections were reacted with the normal mouse
serum as a replacement for the primary TSP1 antibody, the
positive staining for TSP1 was abolished (Fig. 6F).
Discussion
TSP1, a glycoprotein that is structurally and functionally
similar to the adhesive glycoproteins, has been shown to
modulate the attachment of a variety of cell types in vitro
(24 –26). Several integrin receptors have been reported to
bind TSP1 (26, 27), indicating its role in cell-matrix interactions. It has also been proposed that TSP1 plays a role in
autocrine control of cell growth (28). Furthermore, TSP1 may
be an activator or stabilizer of TGF-b (29).
Although the cell biology of TSP1 has been extensively
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
2369
FIG. 6. In situ hybridization and immunocytochemical localization of TSP1 mRNA and protein in the myometrium of the pregnant sheep. A,
B, and C, In situ localization of TSP1 mRNA in the myometrium from one STL sheep. Silver grains formed by TSP1 cDNA probes hybridized
with TSP1 mRNA are localized on myometrial fibroblasts (A) and myometrial smooth-muscle cells (B) in the myometrium. Note that silver
grains, indicating the specific hybridization of probe to the TSP1 mRNA in myometrium, are much more abundant in fibroblasts (A) than in
the smooth-muscle cells (B). The specific hybridization signals between TSP1 cDNA probe and the TSP1 mRNA were abolished by the digesting
of the myometrial section with RNase A before being subjected to hybridization (C). D, E, F, Immunocytochemical localization of TSP1 protein
in the myometrium from one STL sheep. Note: TSP1 protein was localized in the extracellular matrix [with (D1) or without (D2) hematoxylin
counterstaining], as well as in the smooth-muscle cells (E). There was no staining observed when the TSP1 primary antibody was replaced by
the normal mouse serum (F). Magnification, 5003.
studied at the molecular and cellular levels, changes in TSP1
in critical uterine tissues have not been studied in relation to
pregnancy and parturition in any species. Throughout preg-
nancy, the myometrium is undergoing extensive biochemical
and physiological changes. Myometrial activity initially remains relatively quiescent throughout pregnancy, demon-
2370
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
strating only irregular, long-lasting epochs of activity (contractures). At STL, the myometrium becomes active,
developing regular contractions to expel the fetus (2). The
genes controlling these changes are poorly understood. SSH
is a powerful and sensitive technique for rapidly identifying
changes in abundance of mRNAs for individual genes during physiological events. It is a powerful method for studying the changing pattern of gene expression that allows identification of both known and previously undescribed
mRNAs.
Numerous cDNA subtraction methods have been reported (30 –32). They are usually inefficient for obtaining
low-abundance transcripts, and they require considerable
amounts of poly-(A)1 RNA, involve multiple or repeated
subtraction steps, and are labor intensive (23). The mRNA
differential display (33) and RNA fingerprinting by arbitrary
primed PCR (34) are potentially faster methods for identifying differentially expressed genes. However, both of these
methods have a high level of false positives (35, 36) and are
biased for high copy number mRNA (37).
Our study represents the first use of the SSH technique to
examine transcript changes at the time of parturition. We
demonstrated a marked increase of TSP1 associated with STL
in myometrium collected from the central body of the pregnant uterus. Up-regulation of TSP1 mRNA in the pregnant
sheep myometrium at STL, identified by SSH, was confirmed
by Northern blot analysis. The demonstration that similar
changes in TSP1 gene expression are associated with both
STL and BPL supports the view that increased uterine TSP1
production plays an important role in the uterine changes
that accompany labor, regardless of the manner in which
labor is initiated.
It is interesting to note that inhibition of myometrial contraction by nimesulide resulted in the suppression of myometrial TSP1 expression during labor. The nimesulide infusion regimen that we used resulted in inhibition of
myometrial contractility (38) and significantly inhibited PG
production in the fetal circulation (39). Thus, our findings
provide further evidence of coupling of PG concentration
and the progress of labor with the active synthesis of uterine
labor-related genes, including TSP1. These findings indicate
that uterine activation and stimulation are both required for
the normal progression of labor.
We also report here that the increased TSP1 mRNA is
carried to the protein level. TSP1 mRNA rose concurrently
with TSP1 protein. The molecular weights we observed are
consistent with the report that TSP1 is readily proteolyzed in
tissue lysates, resulting in bands on Western blot of lower
molecular weight (120 or 140K) than the intact 180K subunit
(from the data sheet provided by the TSP1 antibody vendor).
The mechanisms that up-regulate TSP1 production during
labor are not defined. Parturition in sheep is accompanied by
both a rise in estradiol and fall in progesterone in maternal
plasma, together with changes in a large number of other
potential regulators. The involvement of estradiol or/and
progesterone in the regulation of increased TSP1 during labor awaits further study.
Increased TSP1 mRNA in the uterine lower segment, identified by Northern blot analysis, has been found in association with human labor (40). In our study, we were able to
Endo • 1999
Vol 140 • No 5
evaluate changes in the body of the uterus, which plays a
more active role in labor than the lower segment. Differential
molecular mechanisms are associated with the regional control of contractility of the pregnant uterus; for example, inhibitory and contractile PG receptors are differentially distributed in the uterine corpus and fundus, compared with
lower segment (41). Therefore, the information obtained
from uterine lower segment cannot be simply applied to the
uterine corpus and fundus. In the current study, the increase
in TSP1 mRNA was confirmed by three different molecular
approaches: SSH, dot blot analysis, and Northern blot analysis. In addition, we quantitatively demonstrated, by Western blot analysis, that the increased TSP1 mRNA was accompanied by increased TSP1 protein. The function of upregulated myometrial TSP1 during parturition is not clear.
Because TSP1 increases with labor in human myometrium
(40), we hypothesize that TSP1 may act as an indirect regulator of myometrial contractility. Further studies are needed
to clarify the potential function of myometrial TSP1 in myometrial activation.
In the present study, we immunolocalized the TSP1 in the
pregnant sheep myometrium. We validated the TSP1 antiserum, by Western blot, before performing immunocytochemistry, showing that the antiserum specifically hybridized to proteins with expected molecular weights.
Determination of the precise distribution of TSP1 in pregnant
sheep myometrium is essential for a complete understanding
of TSP1’s action in its target tissue. Specific immunostaining
for TSP1 was mainly associated with fibroblasts in the pregnant sheep myometrium. However, myometrial cells also
contained TSP1 localized by immunostaining. To further determine which cell type is responsible for producing TSP1, in
situ hybridization was performed. Using in situ hybridization, TSP1 mRNA was identified mainly in fibroblasts; however, some myometrial cells also contained TSP1 mRNA,
indicating that both fibroblasts and myometrial cells are involved in producing TSP1.
In conclusion: 1) our data provide the first evidence that
changes in TSP1 mRNA and protein are associated with
myometrial activation during both STL and BPL in sheep; 2)
both myometrial fibroblasts and smooth muscle cells are
responsible for producing TSP1; and 3) SSH is a powerful
technique that enables us to study differentially regulated
genes during labor.
Acknowledgment
We would like to thank Dr. Gordon Smith for his helpful comments
on the manuscript.
References
1. Lye SL 1994 The initiation and inhibition of labor-toward a molecular understanding. Semin Reprod Endocrinol 12:284 –297
2. Nathanielsz PW, Bailey A, Poore ER, Thorburn GD, Harding R 1980 The
relationship between myometrial activity and sleep state and breathing in fetal
sheep throughout the last third of gestation. Am J Obstet Gynecol 138:653– 659
3. Harding R, Poore ER, Bailey A, Thorburn GD, Jansen CAM, Nathanielsz PW
1983 Electromyographic activity of the nonpregnant and pregnant sheep
uterus. Am J Obstet Gynecol 142:448 – 457
4. Wu WX, Myers DA, Nathanielsz PW 1995 Changes in estrogen receptor
mRNA in sheep fetal and maternal tissues during late gestation and labor. Am J
Obstet Gynecol 172:844 – 850
5. Wu WX, Derks JB, Nathanielsz PW 1996 The effect of glucocorticoids on
TSP1 EXPRESSION IN THE PREGNANT SHEEP MYOMETRIUM
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
estrogen receptor messenger ribonucleic acid in the pregnant ovine myometrium in vivo and in vitro. Biol Reprod 54:230 –241
Wu WX, Derks JB, Zhang Q, Nathanielsz PW 1996 Changes in heat shock
protein 90 and 70 mRNA in uterine tissues of the ewe in relation to parturition
and regulation by estradiol and progesterone. Endocrinology 137:5685–5693
Lefebvre DL, Giaid A, Bennett H, Lariviere R, Zingg HH 1992 Oxytocin gene
expression in rat uterus. Science 256:1553–1555
Miller FD, Chibbar R, Mitchell BF 1993 Synthesis of oxytocin in amnion,
chorion and decidua: a potential paracrine role for oxytocin in the onset of
human parturition. Regul Pept 45:247–251
Zingg HH, Rozen F, Chu K, Larcher A, Arslan A, Richard S, Lefebvre D 1995
Oxytocin and oxytocin receptor gene expression in the uterus. Recent Prog
Horm Res 50:255–272
Chen ZQ, Qi S, Lye SL 1996 Myometrial expression and regulation of the rat
oxytocin receptor during term and preterm labour. J Soc Gynecol Invest
[Suppl] 3:A77P
Fuchs AR, Periyasamy S, Alexandrova M, Soloff MS 1983 Correlation between oxytocin receptor concentration and responsiveness to oxytocin in pregnant rat myometrium. Effect of ovarian steroids. Endocrinology 113:742–749
Wu WX, Nathanielsz PW 1994 Changes in oxytocin receptor messenger RNA
in the endometrium, myometrium, mesometrium, and cervix of sheep in late
gestation and during spontaneous and cortisol-induced labor. J Soc Gynecol
Invest 1:191–196
Wu WX, Verbalis JG, Hoffman GE, Derks JB, Nathanielsz PW 1995 Characterization of oxytocin receptor (OTR) expression and distribution in the
pregnant sheep uterus. Endocrinology 137:722–728
Bernal AL, Europe-Finner GN, Phaneuf S, Watson SP 1995 Preterm labour:
a pharmacological challenge. Trends Pharmacol Sci 16:129 –133
Pierce KL, Gil DW, Woodward DF, Regan JW 1995 Cloning of human prostanoid receptors. Trends Pharmacol Sci 16:253–256
Senior J, Marshall K, Sangha R, Clayton JK 1993 In vitro characterization of
prostanoid receptors on human myometrium at term pregnancy. Br J Pharmacol 108:501–506
Garfield RE 1988 Structural and functional studies of the control of myometrial
contractility and labour. In: McNellis D, Challis J, MacDonald P, Nathanielsz
P, Roberts J (eds) The Onset of Labor: Cellular and Integrative Mechanisms.
Perinatology Press, Ithaca NY, pp 55– 80
Lye SL, Nicholson BJ, Mascarenhas M, NacKenzie L, Petrocelli T 1993 Increased expression of connexin - 43 in the rat myometrium during labor is
associated with an increase in the plasma estrogen:progesterone ratio. Endocrinology 132:2380 –2386
Batra S 1985 Increase by estrogen of calcium entry and calcium channel density
in uterine smooth muscle. Br J Pharmacol 85:388 –390
Inoue Y, Speralakis N 1991 Gestational change in Na1 and Ca21 channel
current densities in rat myometrial smooth muscle cells. Am J Physiol
260:C658 –C663
Wu WX, Unno N, Ma XH, Nathanielsz PW, Inhibition of prostaglandin production by nimesulide is accompanied by changes in expression of the “cassette” of uterine labor related genes in the pregnant sheep. Endocrinology, in
press
Figueroa JP, Mahan S, Poore ER, Nathanielsz PW 1985 Characteristics and
analysis of uterine electromyographic activity in the pregnant sheep. Am J
Obstet Gynecol 151:524 –531
Diatchenko L, Lau YFC, Campbell AP, Chenchik A, Moqadam F, Huang B,
Lukyanov S, Lukyanov K, Gurskaya N, Sverdlov ED, Siebert PD 1996 Sup-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
2371
pression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci USA
93:6025– 6030
O’Shea KL, Liu LHJ, Dixit VM 1991 Thrombospondin and a 140-kDa fragment
promote adhesion and neurite outgrowth from embryonic central peripheral
neurons and from PC12 cells. Neuron 7:231–237
Roberts DD, Sherwood JA, Ginsburg V 1987 Platelet thrombospondin mediates attachment and spreading of human melanoma cells. J Cell Biol
104:131–139
Stomski FC, Gani CJ, Bates RC, Burns GF 1992 Adhesion to thrombospondin
by human embryonic fibroblasts is mediated by multiple receptors and includes a role for glycoprotein 88 (CD36). Exp Cell Res 198:85–92
Lawler J, Weinstein R, Hynes RO 1988 Cell attachment to thrombospondin:
the role of arg-gly-asp, calcium, and integrin receptors. J Cell Biol
107:2351–2361
Mosher DF 1990 Physiology of thrombospondin. Annu Rev Med 41:85–97
Munger JS, Harpel JG, Glezes PE, Mazzieri R, Nunes I, Rifkin DB 1997
Latent transforming growth factor-b: structural features and mechanisms of
activation. Kidney Int 51:1376 –1382
Duguid JR, Dinauer MC 1990 Library subtraction of in vitro cDNA libraries
to identify differentially expressed genes in scrapie infection. Nucleic Acids
Res 18:2789 –2792
Hendrick SM, Cohen DL, Nielsen EA, Davis MM 1984 Isolation of cDNA
clones encoding T cell-specific membrane-associated proteins. Nature
308:149 –153
Sargent TL, Dawid IB 1983 Differential gene expression in the Gastrula of
Xenopus laevis. Science 222:135–139
Liang P, Pardee AB 1992 Differential display of eukaryotic messenger RNA by
means of the polymerase chain reaction. Science 257:967–970
Welsh J, Chada K, Dalai SS, Cheng R, Ralph D, McClelland M 1992 Arbitrarily primed PCR fingerprinting of RNA. Nucleic Acids Res 20:4965– 4970
Bauer D, Warthoe P, Rohde L, Struss M 1994 PCR Methods and Applications.
Cold Spring Harbor Laboratory Press, Plain-View NY, pp S97–S108
Sompayac L, Jane S, Burn TC, Tenen DG, Danna KJ 1995 Overcoming
limitations of the mRNA differential display technique. Nucleic Acids Res
23:4738 – 4739
Bertioli DJ, Schichter UHA, Adams MJ, Burrows PR, Steinbiss HH, Antoniw
JF 1995 An analysis of differential display shows a strong bias towards high
copy number mRNAs. Nucleic Acids Res 23:4520 – 4523
Unno N, Jenkins S, Wu WX, Wong CH, Bennett PR, Nathanielsz PW 1997
Effects of maternal administration of nimesulide (NIM), a putative prostaglandin (PG) endoperoxide synthase (PGHS)-2 inhibitor on myometrial (MYO)
contraction power in sheep during spontaneous term labor (STL). J Soc Gynecol [Suppl] 4:A336P
Unno N, Wu WX, Wong CH, Shinozuka N, Nathanielsz PW 1998 Prostaglandin regulation of fetal plasma adrenocorticotropin and cortisol concentrations in late gestation sheep. Biol Reprod 58:514 –519
Morimoto T, Head JR, Casey ML, MacDonald PC 1998 Thrombospondin-1
expression in human myometrium before and during pregnancy, before and
during labor, and in human myometrial cells in culture. J Soc Gynecol Invest
5:F633
Gordon CSS, Baguma-Nibasheka M, Wu WX, Nathanielsz PW 1998 Regional
variation in prostaglandin and receptors in baboon gravid uterus. Am J Obstet
Gynecol 179:1545–1552