BIOLOGY OF REPRODUCTION 62, 1804–1811 (2000)
Horse Conceptuses Secrete Insulin-Like Growth Factor-Binding Protein 31
A. Herrler,2,3,4,5 Jenny M. Pell,3 W.R. Allen,4 H.M. Beier,5 and Francesca Stewart3,4
The Babraham Institute,3 Babraham, Cambridge, CB2 4AT, United Kingdom
University of Cambridge,4 Department of Clinical Veterinary Medicine, Equine Fertility Unit, Newmarket,
Suffolk CB8 9BH, United Kingdom
Department of Anatomy and Reproductive Biology,5 RWTH, Aachen, Germany
ABSTRACT
Insulin-like growth factor-I (IGF-I) promotes early embryonic
development in several species. In the rabbit, IGF-I binds to the
embryonic coats from Day 3 of development onward by a 38kDa protein that is probably insulin-like growth factor-binding
protein 3 (IGFBP3). In the present study, ligand, Western, and
Northern blot analyses were used to demonstrate the presence
of IGF-I-binding activity, several immunoreactive IGFBP3 proteins, and IGFBP3 mRNA in horse conceptuses with particularly
large amounts of immunoreactive IGFBP3 in the conceptus capsule. In addition, immunoprecipitation of radiolabeled proteins
showed that cultured horse conceptuses secreted IGFBP3 into
the culture medium. Endometrial samples from mares also contained IGFBP3 mRNA and protein; but there was no evidence
of secretion of IGFBP3 into the uterine lumen by ligand blot
analysis, and there was evidence of only very small amounts by
Western blot analysis.
These results indicate that the horse conceptus secretes significant quantities of IGFBP3 toward the conceptus capsule from
as early as Day 10 after ovulation. Thus, most of the IGFBP3
contained within the capsule, which binds IGF-I to this special
extracellular matrix of the preimplantation horse conceptus, is
likely to be embryonic in origin. IGFBP3 in the horse conceptus
capsule may enhance or modulate the action of IGFs on the
developing conceptus.
INTRODUCTION
Several growth factors play important roles in early development of the conceptus via their effects on cell proliferation and morphogenesis. Insulin-like growth factor
(IGF)-I and IGF-II have both been implicated in early development [1–3], and in vitro studies have shown that IGFI can stimulate glucose uptake, protein synthesis, cell proliferation, and conceptus development [4–7]. Furthermore,
IGF-I appears to act as a survival factor by decreasing apoptosis [7]. IGF-I receptors are present in mouse embryos
from the compacted 8-cell stage of development and in rabbits from the morula stage onward [8–10]. However, studies
on the synthesis of IGF-I by the conceptus prior to implantation have led to conflicting results. While some investigators were unable to detect IGF-I mRNA in preimplantation embryos [9, 11, 12], others described its presence from
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG-grant, He 2688/1-1, to A.H.) and the Thoroughbred Breeders’ Association of Great Britain.
2
Correspondence: Andreas Herrler, Department of Anatomy and Reproductive Biology, University of Aachen, RWTH, Wendlingweg 2, 52057
Aachen, Germany. FAX: 49 241 88 88 508; e-mail:
[email protected]
1
Received: 8 April 1999.
First decision: 14 May 1999.
Accepted: 31 January 2000.
Q 2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
the oocyte to the blastocyst stage [13, 14]. However, IGFI protein has not been shown to be synthesized by the preimplantation conceptus [11, 15]. On the other hand, maternal IGF-I is present in oviductal and uterine secretions and
therefore surrounds the conceptus at all times [15–17]. We
demonstrated previously that IGF-I is bound to the external
coat of the preimplantation rabbit conceptus by a 38-kDa
protein that is probably insulin-like growth factor-binding
protein 3 (IGFBP3) [10]. In addition, mRNAs for IGFBP
2, 3, and 4 are present in preimplantation mouse and bovine
conceptuses [15, 18]. Together these findings suggest that
preimplantation mammalian embryos synthesize IGFBPs
that may regulate the bioavailability and action of IGFs on
preimplantation development. IGFBP3 appears to be of
special interest in this regard, as in coculture with Vero cells
this binding protein has been implicated as being responsible for the observed enhanced development of embryos
[19, 20].
The coats of conceptuses are extracellular matrices
(ECM) that surround them during most of the time of preimplantation development [21, 22]. This ECM is formed
initially by the oocyte and granulosa cells and is modified
subsequently by oviductal, uterine, and, in some cases, conceptus secretions [21–24]. In the horse, the zona pellucida
surrounds the oocyte and conceptus until Day 8 after ovulation [22]. On Day 6 a new coat, the capsule, forms beneath the zona pellucida [22, 25]. This is derived mainly
from products of the conceptus [24] and is similar to the
neozona in the rabbit [21]. The horse capsule increases approximately 20-fold in mass between Days 10 and 18 after
ovulation, but it begins to disintegrate at about Day 20 and
has disappeared by Day 23 [26, 27].
To investigate IGFBPs in the horse conceptus, early conceptuses (Days 10–16) were recovered nonsurgically from
mares and cultured in serum-free medium. The conceptus,
capsule, and culture medium were analyzed for the presence of IGFBPs by ligand blotting using 125I-IGF-I and by
Western blotting using an anti-IGFBP3 serum. In addition,
tissues (liver, endometrium, conceptus) were analyzed by
Northern blotting to detect mRNA for IGFBP3, and the
supernatants of conceptuses cultured in the presence of
[35S]methionine were analyzed by immunoprecipitation
with the anti-IGFBP3 serum.
MATERIALS AND METHODS
Sample Collection
Twelve thoroughbred mares, aged 3–7 yr, were used,
some of them on more than one occasion, for the collection
of 20 early conceptuses (Days 10–16) and one Day 30 conceptus. Follicular development and ovulation were monitored by daily transrectal ultrasound examination of the
ovaries, and the mares were inseminated once with diluted
fresh stallion semen when they showed a follicle of .35-
1804
IGFBP3 IN HORSE CONCEPTUSES
mm diameter. Ovulation was confirmed subsequently by
measuring a rise in progesterone concentrations in peripheral serum recovered daily, during and after estrus [28].
Day 1 of pregnancy was taken as the first day on which
the serum progesterone concentration exceeded 1 ng/ml.
Pregnancy was confirmed, and the conceptus was assessed
for regular development by transrectal ultrasound examination from Day 10 onward.
Immediately before the conceptus was removed, a sample of uterine secretion was obtained by infusion of 20 ml
sterile PBS into the horn of the uterus that did not contain
the conceptus via a catheter passed through the working
channel of a strobe-lit videoendoscope, as described by
Bracher and Allen [29]. The conceptus migrates from one
horn to the other frequently during this early time of development, and there should therefore be no difference between the two horns. The uterus was massaged gently per
rectum, and as much fluid as possible was removed by directing the tip of the catheter into the pool of fluid. The
conceptus was then recovered intact on Days 10, 12, or 16
of gestation by uterine lavage using an endotracheal tube
passed through the cervix and flushing with 500–1000 ml
sterile PBS. In total, 20 conceptuses were collected (five
Day 10, ten Day 12, five Day 16), washed several times in
PBS, and either cultured for immunoprecipitation or processed for ligand, Western, or Northern blot analysis.
For immunoprecipitation, five Day 10 and four Day 12
conceptuses were cultured individually for 4 h in a mixture
of 1 ml 1:1 Dulbecco’s modified Eagle’s medium (DMEM)
and RPMI medium, without glutamine and methionine
(ICN, Thame, UK), supplemented with 3 mmol glutamine,
30 mmol methionine, and 5.5 MBq [35S]methionine per milliliter (0.13 mmol; 43.5 TBq/mmol; ICN). Before and after
culture, the diameter of each conceptus was measured and
its volume estimated in order to assess conceptus viability
and growth during culture. A stable or increasing conceptus
volume is a valid measure of viability and growth in the
horse, as the conceptus fluid at this developmental stage is
hypotonic compared to culture media [30] and if damaged,
the conceptus shrinks inside the capsule due to loss of water
from the conceptus cavity toward the culture medium. The
medium was collected and stored at 2708C prior to analysis.
For ligand and Western blot analysis, six Day 12 and
three Day 16 freshly harvested conceptuses were washed
several times in PBS and once in sterile water. Each conceptus was placed on a dry glass microscope slide from
which excess fluid was removed with a thin pipette before
it was punctured with a glass needle and the conceptus fluid
collected. The residual conceptus was then floated in a drop
of water so that the conceptus membranes and capsule
could be collected separately and stored at 2708C until they
were analyzed.
For Northern blot analysis, two of the Day 16 freshly
harvested conceptuses were used. In addition, a Day 30
conceptus was collected nonsurgically by being punctured
with a sharpened catheter through the working channel of
the videoendoscope before the collapsed conceptus was
flushed from the uterus with PBS.
Samples of the endometrium were collected by taking
biopsies on Day 0 (estrus) and from pregnant mares on
Days 10, 12, and 18 after ovulation (2 samples for each
day). Liver samples were collected from a slaughtered
sheep and horse. All samples were frozen immediately in
liquid nitrogen and stored at 2708C until required for RNA
extraction.
1805
Ligand Blot Analysis
Ligand blots were performed by the method of Hossenlopp et al. [31] using two uterine flushes from each stage
(Days 10, 12, and 16); two endometrial biopsies from each
stage (Days 10, 12, and 16); the capsules, conceptus membranes, and conceptus fluid samples from a Day 12 and a
Day 16 conceptus; and two culture supernatants (Days 10
and 12; see below). The samples were subjected to SDSPAGE using a 5% stacking gel and an 8–18% resolving gel
under nonreducing conditions. The endometrial biopsies,
capsules, and conceptus membranes were homogenized in
approximately 100 ml PBS (0.01 M, pH 7.2)/mg tissue and
diluted in loading buffer (0.5 M Tris, pH 6.8; 10% [v:v]
glycerol; 10% [w:v] SDS; 0.05% [w:v] bromophenol blue),
whereas uterine flushings, conceptus fluid, and culture supernatants were diluted in loading buffer. The samples were
heated at 1008C for 5 min and microcentrifuged, and 20 ml
of each sample was loaded onto the gel. After electrophoresis, the gel was blotted onto a polyvinylidene fluoride
membrane (Sartorius, Epsom, UK) by wet blotting (Trans
Blot Cell; BioRad, Hempstead, UK; transfer buffer: 48 mM
Tris, 3 mM glycine, 20% [v:v] methanol, 1.3 mM SDS, pH
8.3) for 3 h at 200 mA, and the membrane was dried at
room temperature for 10 min. The blot was then soaked for
30 min in 100 ml saline (65 mM Tris, pH 7.4; 0.05% Nonidet P-40) at 378C and washed for 2 h in 65 mM Tris containing 3% BSA (insulin and IGF-I free) at 48C, and subsequently for 20 min in 65 mM Tris, pH 7.4, containing
150 mM NaCl and 0.2% Tween 20 at 48C. It was then
incubated overnight at 48C with approximately 500 000
cpm 125I-human (h)IGF-I (iodinated using chloramine T to
a specific activity of 50 mCi/mg) in 25 ml 65 mM Tris, pH
7.4, containing 3% BSA and 0.2% Tween 20; next it was
washed twice, each for 15 min, in 100 ml 65 mM Tris, pH
7.4, supplemented with 150 mM NaCl and 0.1% Tween at
room temperature and subsequently three times, each for
15 min, in 100 ml Tris containing 150 mM NaCl. After
drying at room temperature, the blot was exposed to x-ray
film at 2708C (X-Omat AR; IBI-Kodak Ltd., Cambridge,
UK) for 6–9 days.
Western Blot Analysis
The IGFBP3 antiserum used for Western blotting and
immunoprecipitation was raised in a sheep against a synthetic peptide corresponding to residues 67–79 of hIGFBP3
(67CQPSPDEARPLOA79), an antigenic region selected
from computer-generated predictions of residue surface
probability, hydrophilicity, flexibility, and Jameson-Wolf
antigenic index. The peptide was synthesized and the antiserum raised by the Microchemical Facility at the Babraham Institute, Cambridge. Specific binding to recombinantderived nonglycosylated hIGFBP3 (Celltrix Pharmaceuticals, Santa Clara, CA) and to horse serum was tested by
Western blotting (see Fig. 2).
The procedure for the Western blot analysis was similar
to that described previously for the ligand blots, except that
the loading buffer contained 5% (v:v) b-mercaptoethanol;
30 mM Tris, pH 6.8; 1.5 M urea; 7.5% (v:v) glycerol; 0.5%
(w:v) SDS; and 0.05% (w:v) bromophenol blue. After blotting, the membrane was blocked with 5% (w:v) milk powder in PBS-Tween 20 (0.1% w:v) for at least 1 h at room
temperature. The primary antibody or preimmune serum
was diluted 1:2000 in 5% milk powder in PBS and incubated with the blot at 48C overnight. After three washes in
PBS/Tween, the second antibody (rabbit anti-sheep IgG la-
1806
HERRLER ET AL.
beled with horseradish peroxidase diluted 1:5000 in PBS/
Tween; Dako Ltd., Glostrup, Denmark) was added and left
at room temperature for 1 h. After further rigorous washing,
detection of the second antibody was performed using ECL
reagents (Amersham, Little Chalfont, UK; ECL: horseradish peroxidase-induced chemiluminescent reaction) and exposure to x-ray film (X-Omat AR; IBI-Kodak Ltd.) for 1–
3 min.
Northern Blot Analysis
Total RNA was isolated from two intact Day 16 conceptuses; one Day 30 conceptus; six endometrial biopsies
taken at Days 10, 12, and 18 of gestation (2 at each stage);
and the horse and sheep liver samples using the method of
Chomczynski and Sacchi [32]. Approximately 20 mg total
RNA per track was loaded onto a 1% (w:v) agarose formaldehyde gel, which was then run for 3 h at 50 V. After the
gel was soaked in a denaturing solution of 0.05 M NaOH
and 0.15 M NaCl for 15 min followed by rinsing in 1 M
Tris, pH 7, containing 0.15 M NaCl for 15 min, the RNA
was blotted onto a nylon membrane (Hybond N; Amersham) using positive pressure. The RNA was cross-linked
to the membrane by UV light (Stratagene Crosslinker; Stratagene, Cambridge, UK), and the membrane was prehybridized in approximately 8 ml QuickHyb buffer (Stratagene)
at 608C for 1–2 h. The IGFBP3 probe was a cloned 285base pair polymerase chain reaction product generated from
nucleotides 400–685 of bovine IGFBP3 and checked by
DNA sequence analysis. Approximately 50 ng of the cloned
fragment was labeled with [32P]dCTP (ICN) by random
priming, separated from unincorporated [32P]dCTP using a
NucTrap column (Stratagene), and allowed to hybridize
with the blot at 608C for 2 h in QuickHyb buffer (Stratagene). The membrane was washed twice for 10 min at room
temperature in double-strength SSC (single-strength SSC is
0.15 M sodium chloride and 0.015 M sodium citrate), 0.1%
(w:v) SDS, and once in single-strength SSC, 0.1% (w:v)
SDS for 15 min at 608C; it was then exposed to x-ray film
(X-Omat; IBI-Kodak Ltd.) for 16–48 h.
FIG. 1. This representative blot shows IGFBPs detected by ligand blot
in lane A: endometrium, pregnant, Day (d)12; lane B: uterine flushing,
pregnant, d12; lane C: capsule of conceptus, d12; lane D: conceptus,
d12; lane E: conceptus, d16; lane F: conceptus fluid of d12; lane G:
conceptus fluid of d16; lane H: culture medium in which a d12 conceptus
had been incubated for 4 h. The major protein showing binding to 125IIGF-I was about 46 kDa in size and was strongest in the capsule from
d12 conceptuses.
sheets of cellophane for 36 h and exposed to x-ray film (XOmat AR; IBI-Kodak Ltd.) for 3 wk.
The range of proteins synthesized by the embryo and
secreted into the culture medium was also analyzed by autoradiography. Aliquots (180 ml) of pooled culture media
were freeze-dried on a Speedi Vac (Savant, Halbrook, NY),
resuspended in 20 ml loading buffer as above, heated to
1008C for 5 min, and loaded onto a 12% gel. After electrophoresis, the gel was fixed and dried as described above
and exposed to x-ray film (X-Omat AR; IBI-Kodak Ltd.)
for 4 days.
RESULTS
Immunoprecipitation
Ligand Blot Analysis
The culture medium supernatants in which the Day 10
or 12 conceptuses had been cultured for 4 h in the presence
of [35S]methionine were pooled to give two samples (Day
10 and Day 12). Aliquots (180 ml) were incubated with 20
ml of the sheep anti-hIGFBP3 serum or a control antiserum
directed against a 19-kDa endometrial protein [33] that accumulates in the horse conceptus capsule and is released
into the culture medium [34, 35]. After incubation on a rotating shelf overnight at 48C, 50 ml of washed protein G
sepharose beads (Pharmacia, St. Albans, UK) was added to
the sample antiserum mixture, and incubation was continued on the rotating shelf for a further 24 h at 48C. After
centrifugation at 7000 3 g for 2 min, the pellet was washed
three times with Tris buffer (50 mM, pH 7.4) containing
0.5% Triton X-100, mixed with loading buffer (125 mM
Tris, pH 6.8; 10% [w:v] SDS; 20% [v:v] glycerol, 0.05%
[w:v] bromophenol blue), and heated to 1008C for 5 min.
After brief centrifugation, 40 ml of the supernatant was
loaded onto a 12% gel that was run for 1.3 h at 80 V. The
proteins were then cross-linked overnight in 40% (v:v)
methanol and 10% (v:v) acetic acid, and the gel was soaked
for 20 min in a solution containing 30% (v:v) ethanol and
4% (v:v) glycerol. Finally, the gel was dried between two
Representative examples of the ligand blot analysis of
the endometrium, the uterine flushes, the capsule, the conceptuses, the conceptus fluid samples, and the culture medium supernatants from the incubation of conceptuses are
shown in Figure 1. In all endometrial samples, i.e., a total
of 4 sampled between 12 and 16 days of gestation, proteins
at 46 and 49 kDa bound IGF-I. In contrast, there was no
evidence of any IGFBPs in a total of 4 uterine flushes (20
mg total protein per slot) taken on Days 12 and 16 after
ovulation (Day 12 shown in Fig. 1). The conceptus capsules
showed a very high level of binding at 46 kDa as well as
to a 36 kDa (Day 12 shown in Fig. 1). However, the Day
12 conceptuses themselves did not show any binding, although binding to a 32- and 46-kDa protein was detected
in conceptus fluid. At Day 16, strong binding to a 46-kDa
protein was detected in the conceptus itself and the conceptus fluid; and in the latter, binding to larger proteins
(mainly around 72 kDa) was visible. The samples of culture
medium in which the Day 10 and Day 12 conceptuses had
been incubated for 4 h in the presence of [35S]methionine
showed evidence of a 46-kDa IGFBP (Day 12 shown in
Fig. 1). To check whether 35S present in these samples penetrated the cling film over 7 days, blots of the samples
IGFBP3 IN HORSE CONCEPTUSES
FIG. 2. This representative Western blot shows proteins detected by an
antiserum against a hIGFBP3 peptide (67CQPSPDEARPCOA79) generated
in sheep. Lane A, 200 ng nonglycosylated recombinant hIGFBP3 and lane
C, 2 ml horse serum from a Day (d)12 pregnant mare probed with the
antiserum at a dilution of 1:2000. As a control, serum of the sheep collected before immunization was used: lanes B and D show the same
samples (200 ng hIGFBP3 and 2 ml horse serum from a d12 mare) probed
with the preimmune serum at a dilution of 1:2000. The antiserum reacted
very well with the recombinant hIGFBP3 (lane A) and also with several
proteins present in horse serum (lane C).
(prior to binding analysis) were exposed to x-ray film under
identical conditions. The films were completely negative.
Western Blot Analysis
The antiserum raised against a synthetic peptide based
on a hIGFBP3 sequence, but not the serum collected before
immunization, readily detected 200 ng nonglycosylated
hIGFBP3 at 32 kDa (lanes A and B, Fig. 2). The same
antiserum (but not the preimmune serum; lane D, Fig. 2)
bound to proteins of 32, 49, 60, and 80 kDa in the serum
of mares on Day 12 of gestation (lane C, Fig. 2). Different
forms of IGFBP3 similar to these have been described previously in other species by Baxter and Martin [36], indicating that the anti-hIGFBP3 serum binds specifically to
horse IGFBP3 isoforms.
Since the culture medium samples, the capsules, the conceptus (Day 16), and the conceptus fluid all showed evidence of a 46-kDa IGFBP (see Fig. 1), Western blot analysis was employed to determine whether immunoreactive
IGFBP3 was present in these samples. In all endometrial
samples taken at Days 12 and 16 of pregnancy, a 28-kDa
protein and two bands at 46 and 49 kDa were detectable
(Day 12 sample shown in lane A, Fig. 3); and although no
IGFBP could be detected by ligand blot in uterine fluid (see
1807
FIG. 3. A representative Western blot using the same hIGFBP3 antiserum
as in Figure 2. Lane A: endometrium, pregnant, Day (d)12; lane B: uterine
flush, d12; lane C: capsule of conceptus, d12; lane D: conceptus, d12;
lane E: conceptus fluid, d12; lane F: culture medium in which a d12
conceptus had been incubated for 4 h. All samples contained immunoreactive proteins, some of which were smaller than the nonglycosylated
recombinant hIGFBP3 (see Fig. 2) and may therefore represent cleaved
products of equine IGFBP3.
Fig. 1), two bands at 46 and 49 kDa and a single band at
20 kDa were visible by Western blot in uterine fluid from
the mares at Days 12 and 16 of gestation (Day 12 sample
shown in lane B, Fig. 3). In the Day 12 and Day 16 capsules, the major signal was in the area between 16 and 20
kDa, possibly representing cleaved forms of IGFBP3, with
minor bands at 32, 38, 46, 49, and 60 kDa. The Day 12
and Day 16 conceptuses themselves gave a similar pattern
(32, 38, 46 kDa), except that there was a weaker signal in
the 22-kDa area and there were two additional very strong
bands at approximately 80 and 150 kDa. The Day 12 conceptus fluid contained low amounts of the 49-kDa protein
and two larger proteins (60 and 150 kDa). Supernatants
from Days 10 and 12 conceptuses cultured for 4 h in the
presence of [35S]methionine showed only a single band at
46 kDa (Day 12 shown in lane F, Fig. 3), which corresponded very well with the 46-kDa protein detected by ligand blot (see Fig. 1).
Northern Blot Analysis
Since the Western blots indicated that IGFBPs in maternal and embryonic tissues detected by ligand blot were likely to be IGFBP3, it was important to determine where
IGFBP3 might be expressed. Northern blot analysis demonstrated a single 2.7-kilobase (kb) transcript in sheep and
horse liver RNA (Fig. 4), which was the expected size for
IGFBP3 mRNA. The RNA from both Day 16 conceptuses
(one shown in lane C, Fig. 4) and the Day 30 conceptus
showed the same 2.7-kb transcript. Although loading of
lane C and D was unequal (see the 18S ribosomal bands in
Fig. 4), the Day 30 embryo appeared to contain more
IGFBP3 mRNA than the Day 16 embryos. Likewise, all of
1808
HERRLER ET AL.
FIG. 4. A representative Northern blot detecting IGFBP3 mRNA in liver
(lane A/sheep; lane B/horse); lane C: conceptus, Day (d)16; lane D: conceptus, d30, and endometrium (lane E/estrus; lane F/pregnant, d10; lane
G/pregnant, d18). A single band of the expected size for IGFBP3 mRNA
(2.7 kB) was detected in all samples. The lower panel shows the 18S
ribosomal RNA band stained with ethidium bromide (before transfer).
the endometrial samples were positive; and although both
Day 16 samples appeared to give a stronger signal than the
other samples, it was not possible to statistically analyze
this because of the small number of samples taken at each
stage. These results indicate that both the conceptus and the
endometrium express IGFBP3 mRNA.
Immunoprecipitation
The conceptuses developed well during 4 h of culture as
reflected by their increase in volume. The Day 10 conceptuses expanded from 49.8 6 17.2 mm3 to 75 6 21.5 mm3
(mean 6 SEM; P , 0.001, n 5 5), and the Day 12 conceptuses from 574.8 6 22.7 mm3 to 609.7 6 32.7 mm3
(mean 6 SEM; P , 0.05, n 5 4).
Since it was not possible to distinguish by Northern blot
analysis where the IGFBP3 present in the conceptus capsule might be synthesized, immunoprecipitation was used
to determine whether the conceptus secreted IGFBP3 into
the culture medium, that is, toward the capsule. Proteins
synthesized by the Day 10 and 12 conceptus were labeled
by being cultured with [35S]methionine, and those proteins
secreted into the culture medium were detected by autoradiography (Day 10 and 12 supernatants shown in lanes A
and B, Fig. 5). After the Day 10 and 12 culture supernatants
were pooled, they were immunoprecipitated with the same
IGFBP3 antibody used in the Western blotting experiments.
Both showed a 48-kDa and a faint 23-kDa band (lanes C
and D, Fig. 5), which were in the expected size range of
IGFBP3 isoforms. The control serum raised against a maternal protein (P19) known to be present in the capsule of
horse conceptuses at this stage [33] gave a negative result
with both samples.
FIG. 5. Lanes A and B show representative autoradiographs of the culture supernatants in which intact horse conceptus had been incubated for
4 h in the presence of [35S]methionine (lane A, Day (d)10 conceptus; lane
B, d12 conceptus). Lanes C, D, and E show the immunoprecipitation results on the same samples using the hIGFBP3 peptide antiserum (lane C,
d10 conceptus; lane D, d12 conceptus) and a control antiserum directed
against P19, another equine protein (lane E, d12 conceptus).
DISCUSSION
These results indicate that the IGFBPs detected by ligand
blotting in early preimplantation horse conceptuses and
their coats are most likely to be isoforms of IGFBP3. Furthermore, mRNA for IGFBP3 was detected in the conceptus, and immunoprecipitable IGFBP3 was shown to be synthesized and secreted by conceptuses as early as Day 10 of
development. Although preimplantation mouse [18] and
cattle conceptuses [15] transcribe IGFBP3, the actual synthesis of IGFBP3 protein by preimplantation conceptuses
has not been shown before.
In a previous study we demonstrated by ligand blot analysis that a 38-kDa IGFBP, likely to be IGFBP3, is present
in the coats of rabbit conceptuses and that the binding of
IGF-I to the coats of rabbit conceptuses increases from the
day of blastocyst formation onward [10]. Furthermore, IGFI receptor mRNA is present from the 8-cell stage onward
[9], and binding of IGF-I has been shown to occur from
the morula stage onward [8, 10]. The question whether
IGF-I is synthesized by preimplantation embryos is controversial. While some investigators were unable to detect
IGF-I mRNA in preimplantation embryos [9, 11, 12], others described its presence from the oocyte to the blastocyst
stage [13, 14]. However, IGF-I protein has not been shown
to be synthesized by the preimplantation conceptus [11,
15]. Therefore, since preimplantation embryos in a number
of species are bathed in maternal IGF-I [16, 17, 37, 38], we
and others have suggested that maternally derived IGF-I
may be regulated by embryonic IGFBPs to support the development of preattachment conceptuses [10, 19].
Whether IGFBP3 potentiates or inhibits IGF-I action has
also led to controversy in the literature (see [39] for re-
IGFBP3 IN HORSE CONCEPTUSES
view). Several reports have described an inhibitory action
for IGFBP3 [40, 41], but potentiation of IGF-I action has
also been demonstrated [42], especially in in vivo studies
[43, 44]. One reason for this controversy may be the higher
affinity that IGFBP3 has for IGF-I in solution compared to
IGFBP3 bound to extracellular matrices or cells. IGFBP3
in solution has a significantly higher affinity for IGF-I than
does the IGF-I receptor and it is clearly capable of preventing IGF-I-receptor interaction. The affinity of IGFBP3
for IGF-I when it is associated with extracellular matrices
or cells is significantly reduced [40, 45], and it is therefore
capable of modulating the amount of IGF-I available for
receptor binding [46]. Also, the local microenvironment is
clearly an important modulator of IGFBP3 action. Conover
et al. [47] demonstrated that, in a neutral environment,
IGFBP3 is capable of blocking IGF-I-stimulated 3H-aminobutyric acid uptake, while in an acidic environment
IGFBP3 potentiates the action of IGF-I. In this context it
should be noted that conceptus coats are acidic because of
their high amount of sialic acid [21, 24]. Therefore,
IGFBP3 in this special ECM may sequester maternal IGFI and potentiate its action as it passes it on to the embryo.
It has been suggested that IGFBP3 may contribute to the
supportive effect of somatic cell coculture on early embryo
development. Vero cells, which promote early embryonic
development in coculture [19], release IGFBP3 into the culture medium [20], and the culture medium containing the
most IGFBP3 was the most effective at promoting development. In addition, granulosa cells [48, 49], oviductal and
endometrial cells [20, 50], and liver cells [51, 52] promote
development of conceptuses in vitro, and these cells all
secrete IGFBP3 [15, 20, 53–56]. In all these coculture systems, IGFBP3 may be one of the factors that supports development. Furthermore, IGFBP3 produced by the embryo
itself may support early embryonic development. It has
been established by Wiley et al. [57] and others [58, 59]
that embryos should be cultured in small groups in small
quantities of culture medium, suggesting the possible existence of autocrine factor(s) supporting development.
IGFBP3 may be such a factor, as it is produced by the
embryo and it supports early embryonic development.
Recently, IGFBP3 fragments, especially those ,20 kDa
in size, have engendered special interest [47]. They lose
their ability to bind IGF-I, but nevertheless continue to be
able to influence the mitogenic effect of IGF-I [60]. This
action independent of IGF-I binding is explained by the
ability of IGFBP3 to bind to cell membranes [61]. A receptor for IGFBP3 has been postulated for some time, and
there is now good evidence for such a receptor [62, 63]. In
this study we provide evidence that the horse capsule contains more of the cleaved forms of IGFBP3 than the native
forms. Lee et al. [64] recently described the presence of
IGFBP3 protease activity within porcine uterine fluid on
Days 11 and 12 of pregnancy. Such a protease could be
responsible for the large amount of cleaved IGFBP3 within
the embryonic coats, which would result in the release of
large amounts of IGFs, thus promoting development of the
conceptus. As the immunoprecipitation showed, only a very
small amount of cleaved IGFBP3 appears to pass through
the capsule and into the culture medium, suggesting that it
may be bound to the capsule. IGFBP3 cleaved by matrix
metalloproteases (MMP1 to MMP3) results in a 16-kDa
fragment (residues 1–176) containing an amino acid sequence (149KKGHA153) that makes it capable of binding to
glycosaminoglycans [65, 66]. Thus it is possible that
cleaved forms of IGFBP3 accumulate within the coats. As
1809
the cleaved forms induce cell death [60], their presence in
the coats could create a defensive wall against attacking
cells.
In addition to the smaller, cleaved fragments, several
high molecular weight IGFBP3-immunoreactive proteins
were detected in the conceptus membranes and conceptus
fluid samples (see Fig. 3). These may represent SDS-stable
multimers of IGFBP3 and/or its cleaved fragments, which
have been shown to form under certain conditions in both
IGFBP3 and IGFBP5, but not IGFBP1 [67]. Furthermore,
the Western blots appear to be much more sensitive than
the ligand blots, particularly for the Day 12 uterine fluid
and conceptus membrane samples, which gave negative results on the ligand blots but several strong bands on the
Western blots. This could be due to a genuine difference in
sensitivity (e.g., differences in gel running conditions, binding constants, and detection methods) but also due to lack
of binding of the labeled IGF-I to some of the immunoreactive fragments and higher molecular weight forms of
IGFBP3. Significantly, only the 46- to 49-kDa protein was
detected via both ligand and Western blotting in the culture
medium samples and also in the immunoprecipitation experiments, apart from a very faint band at 23 kDa. This
represents the protein synthesized over 4 h by the embryo
in culture and indicates that neither the cleavage nor large
complexes of IGFBP3 were able to form under these conditions. In addition, since the size range agrees with the
size range of IGFBP3 in other species, the results provide
very convincing evidence that the equine conceptus secretes IGFBP3.
In summary, the horse conceptus capsule contains
IGFBP3, which, since it is secreted in relatively large
amounts by conceptuses from Day 10 onward, is likely to
be largely embryonic in origin. This might regulate the influence of IGFs on the conceptus. Furthermore, a large
amount of IGFBP3 within the capsule is in the form of
cleaved fragments, possibly the result of proteases, which
would release large amounts of IGFs, thus promoting the
development of the conceptus. The cleaved IGFBP3 forms
a defensive wall against attacking cells, whereas the intact
IGFBP3 might concentrate maternal IGF-I in the capsule
and present it to the embryo.
ACKNOWLEDGMENTS
We thank W. John Coadwell for the protein sequence analysis and
design of the peptide used to raise the IGFBP3 antiserum. We also thank
Tom A.E. Stout and Cornelia Gerstenberg for technical advice and support
during embryo and tissue collections.
REFERENCES
1. Schultz GA, Heyner S. Growth factors in preimplantation mammalian
embryos. Oxf Rev Reprod Biol 1993; 15:43–81.
2. Giudice LC, Saleh W. Growth factors in reproduction. Trends Endocrinol Metab 1995; 6:60–69.
3. Simmen RC, Green ML, Simmen FA. IGF system in periimplantation
uterus and embryo development. In: Dey SK (ed.), Molecular and
Cellular Aspects of Preimplantation Processes. Serono Symposium.
New York: Springer Verlag; 1995: 185–204.
4. Harvey MB, Kaye PL. Visualization of insulin receptors on mouse
pre-embryos. Reprod Fertil Dev 1992; 3:9–15.
5. Rappolee DA, Sturm KS, Behrendtsen O, Schultz GA, Pedersen RA,
Werb Z. Insulin-like growth factor II acts through an endogenous
growth pathway regulated by imprinting in early mouse embryos.
Genes Dev 1992; 6:939–952.
6. Pantaleon M, Kaye PL. IGF-I and insulin regulate glucose transport
in mouse blastocysts via IGF-I receptor. Mol Reprod Dev 1996; 44:
71–76.
7. Herrler A, Krusche CA, Beier HM. Members of the insulin family
1810
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
HERRLER ET AL.
promote rabbit blastocyst development and prevent apoptosis. Biol
Reprod 1998; 59:1302–1310.
Mattson BA, Rosenblum IY, Smith RM, Heyner S. Autoradiographic
evidence for insulin and insulin-like growth factor binding to early
mouse embryos. Diabetes 1988; 37:585–589.
Rappolee DA, Sturm KS, Schultz GA, Pederson RA, Werb Z. The
expression of growth factor ligand and receptors in preimplantation
embryos. In: Heyner S, Wiley LM (eds.), Early Embryo Development
and Paracrine Relationships. UCLS Symposia on Molecular and Cellular Biology, New Series 117. New York: AR Liss; 1990: 11–25.
Herrler A, Einspanier R, Beier HM. Binding of IGF-I to preimplantation rabbit embryos and their coats. Theriogenology 1997; 47:1595–
1607.
Heyner S, Smith RM, Schultz GA. Temporary regulated expression
of insulin and insulin-like growth factors and their receptors in early
mammalian development. Bioassays 1989; 11:171–176.
Lighten AD, Hardy K, Winston RM, Moore GE. Expression of mRNA
for the insulin-like growth factors and their receptors in human preimplantation embryos. Mol Reprod Dev 1997; 47:134–139.
Watson AJ, Hogan A, Hahnel A, Wiemer KE, Schultz GA. Expression
of growth factor ligand and receptor genes in the preimplantation bovine embryo. Mol Reprod Dev 1992; 31:87–95.
Doherty AS, Temeles GL, Schultz RM. Temporal pattern of IGF-I
expression during mouse preimplantation embryogenesis. Mol Reprod
Dev 1994; 37:21–26.
Winger QA, de los Rios P, Han VK, Armstrong DT, Hill DJ, Watson
AJ. Bovine oviductal and embryonic insulin-like growth factor binding proteins: possible regulators of ‘‘embryotrophic’’ insulin-like
growth factor circuits. Biol Reprod 1997; 56:1415–1423.
Herrler A, Hegele-Hartung C, Beier HM. Immunohistological detection of IGF-I in rabbit oviducts and uteri during early pregnancy. Acta
Endocrinol Suppl 1992; 126(suppl 4):79 (abstract).
Carlson B, Hillensjoe T, Nisson A, Toernell J, Billig H. Expression
of insulin-like growth-factor I (IGF-I) in the rat fallopian tube: possible autocrine and paracrine action of fallopian tube-derived IGF-I
on the fallopian tube and on the embryo. Endocrinology 1993; 133:
2031–2039.
Hahnel A, Schultz GA. Insulin-like growth factor binding proteins are
transcribed by preimplantation mouse embryos. Endocrinology 1994;
134:1956–1959.
Menezo YJ, Guerin JF, Czyba JC. Improvement of human early embryo development in vitro by coculture on monolayers of Vero cells.
Biol Reprod 1990; 42:301–306.
Lai YM, Wang HS, Lee CL, Lee JD, Huang HY, Chang FH, Lee JF,
Soong YK. Insulin-like growth factor-binding proteins produced by
Vero cells, human oviductal cells and human endometrial cells, and
the role of insulin-like growth factor-binding protein-3 in mouse embryo co-culture systems. Hum Reprod 1996; 11:1281–1286.
Denker HW, Gerdes HJ. The dynamic structure of rabbit blastocyst
coverings. I. Transformation during regular preimplantation development. Anat Embryol 1979; 157:15–34.
Betteridge KJ. The structure and function of the equine capsule in
relation to embryo manipulation and transfer. Equine Vet J 1989;
8(suppl):92–100.
Leiser R, Denker HW. The dynamic structure of rabbit blastocyst coverings. II. Ultrastructural evidence for a role of the trophoblast in
neozona formation. Anat Embryol 1988; 179:129–134.
Oriol JG, Sharom FJ, Betteridge KJ. Developmentally regulated
changes in the glycoproteins of the equine embryonic capsule. J Reprod Fertil 1993; 99:653–664.
Bonnet R. Die Eihäute des Pferdes. Verh Anat Ges 1889; 3:17–38.
Betteridge KJ, Eaglesome MD, Mitchell D, Flood PF, Beriault R. Development of horse embryos up to twenty two days after ovulation:
observations on fresh specimens. J Anat 1982; 135:191–209.
Enders AC, Liu IK. Lodgement of the equine blastocyst in the uterus
from fixation through endometrial cup formation. J Reprod Fertil
Suppl 1991; 44:427–438.
Allen WR, Mathias S, Lennard SN, Greenwood RE. Serial measurement of peripheral oestrogen and progesterone concentrations in oestrous mares to determine optimum mating time and diagnose ovulation. Equine Vet J 1995; 27:460–464.
Bracher V, Allen WR. Videoendoscopic evaluation of the mare’s uterus: I. Findings in normal fertile mares. Equine Vet J 1992; 24:274–
278.
Waelchli RO, Betteridge KJ. Osmolality of equine blastocyst fluid
from day 11 to day 25 of pregnancy. Reprod Fertil Dev 1996; 8:981–
988.
31. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M.
Analysis of serum insulin-like growth factor binding proteins using
western blotting: use of the method for titration of the binding proteins
and competitive binding studies. Anal Biochem 1986; 154:138–143.
32. Chomczynski P, Sacchi N. Single-step method of RNA isolation by
acid guanidinium thiocyanate-phenol-chloroform extraction. Anal
Biochem 1987; 162:156–159.
33. Crossett B, Suire S, Herrler A, Allen WR, Stewart F. Transfer of a
uterine lipocalin from the endometrium of the mare to the developing
equine conceptus. Biol Reprod 1998; 59:483–490.
34. Stewart F, Charleston B, Crossett B, Barker PJ, Allen WR. A novel
uterine protein that associates with the embryonic capsule in equids.
J Reprod Fertil 1995; 105:65–70.
35. Crossett B, Allen WR, Stewart F. A 19 kDa protein secreted by the
endometrium of the mare is a novel member of the lipocalin family.
Biochem J 1996; 320:137–143.
36. Baxter RC, Martin JL. Binding proteins for the insulin-like growth
factors: structure, regulation and function. Prog Growth Factor Res
1989; 1:49–68.
37. Geisert RD, Lee CY, Simmen FA, Zavy MT, Fliss AE, Bazer FW,
Simmen RC. Expression of messenger RNAs encoding insulin-like
growth factor-I, -II, and insulin-like growth factor binding protein-2
in bovine endometrium during the estrous cycle and early pregnancy.
Biol Reprod 1991; 45:975–983.
38. Ko Y, Lee CY, Ott TL, Davis MA, Simmen RC, Bazer FW, Simmen
FA. Insulin-like growth factors in sheep uterine fluids: concentrations
and relationship to ovine trophoblast protein-1 production during early
pregnancy. Biol Reprod 1991; 45:135–142.
39. Jones JI, Clemmons DR. Insulin-like growth factors and their binding
proteins: biological actions. Endocr Rev 1995; 16:3–34.
40. Liu L, Delbe J, Blat C, Zapf J, Harel L. Insulin-like growth factor
binding protein (IGFBP-3), an inhibitor of serum growth factors other
than IGF-I and -II. J Cell Physiol 1992; 153:15–21.
41. Valentinis B, Bhala A, DeAngelis T, Baserga R, Cohen P. The human
insulin-like growth factor (IGF) binding protein-3 inhibits the growth
of fibroblasts with a targeted disruption of the IGF-I receptor gene.
Mol Endocrinol 1995; 9:361–367.
42. Conover CA, Powell DR. Insulin-like growth factor (IGF)-binding
protein-3 blocks IGF-I-induced receptor down-regulation and cell desensitization in cultured bovine fibroblasts. Endocrinology 1991; 129:
710–716.
43. Sommer A, Maack CA, Spratt SK, Masccarenhas D, Tressel TJ,
Rhodes ET, Lee R, Roumas M, Tatsuno GP, Flynn JA, Gerber N,
Taylor J, Cudny H, Nanney L, Hunt TK, Spencer EM. Molecular
genetics and action of recombinant insulin like growth factor binding
protein-3. In: Spencer EM (ed.), Modern Concepts of Insulin-Like
Growth Factors. New York: Elsevier; 1991: 715–728.
44. Clark RG, Mortensen D, Reifsynder D, Mohler M, Etcheverry T,
Mukku V. Recombinant human insulin-like growth factor binding protein-3 (rhIGFBP-3): effects on the glycemic and growth promoting
activities of rhIGF-1 in the rat. Growth Regul 1993; 3:50–52.
45. McCusker RH, Camacho-Hubner C, Bayne ML, Cascieri MA, Clemmons DR. Insulin-like growth factor (IGF) binding to human fibroblast and glioblastoma cells: the modulating effect of cell released
IGF binding proteins (IGFBPs). J Cell Physiol 1990; 144:244–253.
46. McCusker RH, Busby WH, Dehoff MH, Camacho-Hubner C, Clemmons DR. Insulin-like growth factor (IGF) binding to cell monolayers
is directly modulated by the addition of IGF-binding proteins. Endocrinology 1991; 129:939–949.
47. Conover CA, Clarkson JT, Bale LK. Factors regulating insulin-like
growth factor binding protein-3 binding, processing, and potentiation
of insulin-like growth factor action. Endocrinology 1996; 137:2286–
2292.
48. Goto K, Kajihara Y, Kosaka S, Koba M, Nakanishi Y, Ogawa K.
Pregnancies after co-culture of cumulus cells with bovine embryos
derived from in-vitro fertilization of in-vitro matured follicular oocytes. J Reprod Fertil 1988; 83:753–758.
49. Herrler A, Lucas-Hahn A, Niemann H. Effects of insulin-like growth
factor-I on in-vitro production of bovine embryos. Theriogenology
1992; 37:1213–1224.
50. Sakkas D, Trounson AO. Co-culture of mouse embryos with oviduct
and uterine cells prepared from mice at different days of pseudopregnancy. J Reprod Fertil 1990; 90:109–118.
51. Goto K, Iwai N, Takuma Y, Nakanishi Y. Co-culture of in vitro fertilized bovine embryos with different cell monolayers. J Anim Sci
1992; 70:1449–1453.
52. Kobayashi M, Hirako M, Minato Y, Sasaki K, Horiuchi R, Domeki I.
IGFBP3 IN HORSE CONCEPTUSES
53.
54.
55.
56.
57.
58.
59.
Rat hepatoma Reuber H-35 cells produce factors that promote the
hatching of mouse embryos cultured in vitro. Biol Reprod 1997; 56:
1041–1049.
Gargosky SE, Walton PE, Wallace JC, Ballard FJ. Characterization of
insulin-like growth factor-binding proteins in rat serum, lymph, cerebrospinal and amniotic fluids, and in media conditioned by liver, bone
and muscle cells. J Endocrinol 1990; 127:391–400.
Mondschein JS, Smith SA, Hammond JM. Production of insulin-like
growth factor binding proteins (IGFBPs) by porcine granulosa cells:
identification of IGFBP-2 and -3 and regulation by hormones and
growth factors. Endocrinology 1990; 127:2298–2306.
Ko Y, Simmen RCM, Lee CY, Simmen FA. Reversion of the nontransformed phenotype of tsSV40-transformed uterine endometrial epithelial cells is accompanied by decreased mitogenic responsiveness
to IGFs and by enhanced secretion of at least five different IGF binding proteins. Endocr J 1994; 2:495–504.
Liu HC, Mele C, Catz D, Noyes N, Rosenwaks Z. Production of
insulin-like growth factor binding proteins (IGFBPs) by human endometrial stromal cell is stimulated by the presence of embryos. J
Assist Reprod Genet 1995; 12:78–87.
Wiley LM, Yamami S, Van Muyden D. Effect of potassium concentration, type of protein supplement, and embryo density on mouse
preimplantation development in vitro. Fertil Steril 1986; 45:111–119.
Paria BC, Dey SK. Preimplantation embryo development in vitro:
cooperative interactions among embryos and role of growth factors.
Proc Natl Acad Sci USA 1990; 87:4756–4760.
Lane M, Gardner DK. Effect of incubation volume and embryo den-
60.
61.
62.
63.
64.
65.
66.
67.
1811
sity on the development and viability of mouse embryos in vitro. Hum
Reprod 1992; 7:558–562.
Rechler MM. Growth inhibition by insulin-like growth factor (IGF)
binding protein-3—what’s IGF got to do with it? Endocrinology 1997;
138:2645–2647.
Lalou C, Lassarre C, Binoux M. A proteolytic fragment of insulinlike growth factor (IGF) binding protein-3 that fails to bind IGFs
inhibits the mitogenic effects of IGF-I and insulin. Endocrinology
1996; 137:3206–3212.
Oh Y, Muller HL, Pham H, Rosenfeld RG. Demonstration of receptors
for insulin-like growth factor binding protein-3 on Hs578T human
breast cancer cells. J Biol Chem 1993; 268:26045–26048.
Leal SM, Liu Q, Huang SS, Huang JS. The type V transforming
growth factor beta receptor is the putative insulin-like growth factorbinding protein 3 receptor. J Biol Chem 1997; 272:20572–20576.
Lee CY, Green ML, Simmen RC, Simmen FA. Proteolysis of insulinlike growth factor-binding proteins (IGFBPs) within the pig uterine
lumen associated with peri-implantation conceptus development. J Reprod Fertil 1998; 112:369–377.
Fowlkes JL, Suzuki K, Nagase H, Thrailkill KM. Proteolysis of insulin-like growth factor binding protein-3 during rat pregnancy: a role
for matrix metalloproteinases. Endocrinology 1994; 135:2810–2813.
Fowlkes JL, Enghild JJ, Suzuki K, Nagase H. Matrix metalloproteinases degrade insulin-like growth factor-binding protein-3 in dermal
fibroblast cultures. J Biol Chem 1994; 269:25742–25746.
Koedam JA, Hoogerbrugge CM, Van Buul-Offers SC. Insulin-like
growth factor binding proteins-3 and -5 form sodium dodecyl sulfatestable multimers. Biochem Biophys Res Commun 1997; 240:707–
714.